1. Introduction
The occurrence of multiple cyclones in succession of one another is often referred to as a cluster or family of cyclones (Bjerknes and Solberg 1922; Wernli et al. 1999; Fish et al. 2019; Dacre and Pinto 2020; Weijenborg and Spengler 2020), where the first cyclone to develop is sometimes called the “primary” cyclone, and subsequent cyclones that develop upwind and downwind (relative to the mean jet) are referred to as the “upstream” and “downstream” cyclones, respectively. In a baroclinic atmosphere with a zonally continuous waveguide in the upper troposphere, a localized and amplifying disturbance along the jet stream will produce a ridge downstream, followed by a trough farther downstream (Wirth et al. 2018). The formation of these upper-level waves produces a series of alternating surface cyclones and anticyclones through the classical baroclinic development paradigm. While Rossby wave energy dispersion in the upper troposphere is primarily directed downstream, near-surface temperature perturbations can in some cases lead to the bottom-up formation of upstream cyclones and anticyclones (Simmons and Hoskins 1979; Wernli et al. 1999; Wirth et al. 2018). The bottom-up type of upstream development has received less attention in the literature, in part because it is not frequently observed in the real atmosphere. While the existence of this type of upstream development is of interest, there are other ways for upstream cyclones to form and this study is focused on the interaction between pairs of cyclones regardless of how they initially develop.
Air–sea fluxes are known to have substantial influence on the development of extratropical cyclones, provided that there is a sufficient air–sea temperature contrast or vapor pressure deficit. A warmer air temperature relative to the underlying sea surface temperature (SST) will serve to stratify the environment, weaken baroclinicity, and decouple the boundary layer from the free troposphere thereby supporting strong vertical wind shear at the top of boundary layer (Neiman et al. 1990). A cooler air temperature relative to the underlying SST will increase buoyancy within the cyclone through upward surface sensible heat fluxes (SSHF) and thereby deepen the boundary layer, as has been discussed for real extratropical cyclones (Kuo et al. 1991; Gyakum and Danielson 2000; Xu et al. 2011; Booth et al. 2012; Hirata et al. 2015; Phillips and O’Neill 2020) and idealized baroclinic wave simulations (Nuss 1989; Bui and Spengler 2021). A vapor pressure deficit will lead to a moistening of the atmosphere through upward surface latent heat fluxes (SLHF), which can increase latent heating rates leading to a destabilization of the atmosphere and a circulation response associated with the diabatic generation of potential vorticity (PV; Carrera et al. 1999; Demirdjian et al. 2022).
While the impacts of surface fluxes on extratropical cyclone evolution have been extensively investigated, as noted above, less attention has been focused on the role of surface fluxes in strengthening the interaction between clusters of cyclones. Some notable exceptions include Boutle et al. (2010), who found that moisture is sourced into an upstream cyclone’s warm conveyor belt (WCB) and atmospheric river (AR) regions via evaporation from the postfrontal primary cyclone and the anticyclonic boundary layer regions. This is consistent with Sodemann and Stohl (2013), who performed a moisture source analysis on North Atlantic cyclones finding that subsequent cyclones “profit from the moisture transported poleward by previous cyclones, leading to a handover of moisture between subsequent short-lived cyclones.” The airstream described by Boutle et al. (2010) is related to the feeder airstream (FA) termed by Dacre et al. (2019), defined as the low-level airstream that is entrained into a cyclone and feeding into the AR and WCB, the difference being that the Boutle et al. (2010) airstream requires the interaction of cyclone pairs whereas the FA can exist for a solitary cyclone. Papritz et al. (2021) performed a moisture source analysis on hundreds of North Atlantic extratropical cyclone clusters and also found that there is an airstream connection between cyclones within a cluster and that a major moisture source for the precipitation arises from preceding cyclones. It is important to emphasize that it is not only necessary to correctly model the airstream connection between cyclones in a cluster. A model must also capture the preconditioning—warming and moistening prior to storm formation (Booth et al. 2012)—of the airstream as it is handed over from one cyclone to the next to properly predict the cyclone intensity.
The present study is motivated by the desire to 1) better understand the role of SLHF and SSHF in preconditioning the air that contributes toward upstream cyclone development, and 2) identify the moisture source regions that contribute toward upstream cyclone development. This study presents a new analysis that tracks the Boutle et al. (2010) and Dacre et al. (2019) “feeder” airstream over a longer time period to form a more complete picture of how pairs of cyclones can interact. Furthermore, we present a dynamical analysis demonstrating physically how the moistening of the Boutle et al. (2010) airstream can lead to upstream cyclone intensification.
2. Methods
a. Baroclinic wave in a channel
The idealized simulations analyzed in this study are performed with version 5 of the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS; Hodur 1997; Doyle et al. 2014) using the configuration described in Finocchio and Doyle (2019) and Demirdjian et al. (2022). The domain is a 7200 km × 33 000 km latitude–longitude channel from approximately 8° to 72°N in latitude and having periodic boundary conditions in longitude with a horizontal resolution of 30 km. Unlike in Demirdjian et al. (2022), the jet is initialized by inverting a prescribed PV field (Rotunno et al. 1994; Lloveras et al. 2022). This results in a zonally symmetric jet having a maximum westerly wind speed of approximately 52 m s−1 at 40°N, and a width, defined in Manola et al. (2013) as the meridional distance between the inflection points of the meridional jet profile, of approximately 6° (Fig. 1). These values are chosen to be broadly consistent with the probability density functions of jet speed, latitude, and width discussed in Manola et al. (2013). The zonally invariant SST is set to be equal to the initial surface air temperature and stays constant in time while the surface air temperature is allowed to evolve. The initial air temperature is in thermal wind balance with the prescribed wind field and the lowest 500 m is initially set to be dry adiabatic. The jet is perturbed by inverting a prescribed upper-level, localized quasigeostrophic PV field resulting in a balanced meridional and zonal wind anomaly that grows off the baroclinic instability of the jet (Lloveras et al. 2022).
Vertical cross section of the zonal wind speed (color fill; every 5 m s−1) and potential temperature (black lines; every 4 K) of the zonally symmetric jet stream initialized in the idealized simulations.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
b. Air parcel trajectories
The Lagrangian Analysis Tool (LAGRANTO; Sprenger and Wernli 2015) is used to track forward and backward air parcel trajectories using hourly model output. While trajectories computed with finer temporal resolution are not tested here, Demirdjian et al. (2022) did compare LAGRANTO trajectories computed using both hourly and 15-min model output data and found no meaningful difference. The trajectories are computed using a 30-km grid spacing, which is the horizontal grid size of the model. As will be discussed in section 4, trajectories are initialized in the postfrontal sector of the primary cyclone because air parcels in this sector are found to substantially impact the upstream cyclone.
c. Cyclone reference frame
d. Experimental design
The control (CTRL) simulation used in this study is run with full surface fluxes out to 264 h. Several aspects of the CTRL are modified resulting in another six simulations that are the 1) No SF run, where the SSHF and SLHF are switched off at t = 0 h; 2) No SSHF run, where only the SSHF is switched off at t = 0 h; 3) No SLHF run, where only the SLHF is switched off at t = 0 h; 4) No SF 96h run, where the SSHF and SLHF are switched off at t = 96 h; 5) No local SF 96h run, where the SSHF and SLHF are switched off at t = 96 h inside a circle centered on the upstream cyclone with a radius of 8° (890 km); and 6) No nonlocal SF 96h run, where the SSHF and SLHF are switched off at t = 96 h everywhere except a circle centered on the upstream cyclone with a radius of 8° (890 km). The descriptions of the seven simulations used in this study are summarized in Table 1.
List of the simulations used with their names, descriptions, and the simulation time for which the surface fluxes are modified.
3. Synoptic-scale overview
In the CTRL simulation (Figs. 2–4), cyclone development is typical of the life cycle 1 development (Thorncroft et al. 1993), which involves cyclonic Rossby wave breaking on a baroclinically unstable jet (Davies et al. 1991; Thorncroft et al. 1993; Rotunno et al. 1994; Polvani and Esler 2007; Boutle et al. 2010). The three cyclones that form are referred to as the primary cyclone (40°N, 180°), the upstream cyclone (35°N, 145°W), and the downstream cyclone (38°N, 215°W). These three cyclones make up what will be referred to in this study as a cyclone cluster. To more easily track the development of the cyclones, the axes for the figures are relabeled such that the sea level pressure (SLP) minimum of the primary cyclone is always located at 40°N, 180°. The SLP, which is only depicted for approximately half of the full model domain (Fig. 2a), indicates that the primary cyclone begins to develop starting at approximately 120 h into the simulation. By 168 h (Fig. 2b) the SLP minimum of the primary cyclone reaches 996 hPa and a downstream cyclone has begun to form from the downstream propagation of Rossby waves, as seen in the 300-hPa wind speed trough–ridge–trough pattern, which has a SLP minimum of 1004 hPa. At 216 h (Fig. 2c) both the primary and downstream cyclones have matured substantially into cyclones of moderate amplitude having SLP minima of 988 and 980 hPa, respectively. At the same time, the upstream cyclone begins to form at around 35°N, 145°W. By 264 h, (Fig. 2d) the upstream cyclone intensifies further with a SLP minimum of 996 hPa and the primary cyclone continues to grow in size and intensity, while the downstream cyclone mainly grows only in size.
Cyclone-relative plan-view maps every 48 h of the IWV (color fill; every 4 mm), 300-hPa wind speed (green lines; every 5 m s−1), and sea level pressure (black lines; every 4 hPa) for (a) 120, (b) 168, (c) 216, and (d) 264 h. The upstream, primary, and downstream cyclones are labeled in (d). A meridional dashed line in (c) denotes the location of the cross sections in Fig. 7.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
Pressure–longitude cross section at (a) 168, (b) 216, and (c) 264 h of the meridional wind speed (southerly in solid and northerly in dashed; every 4 m s−1) averaged from 30° to 40°N. The locations of the upstream, primary, and downstream cyclones are labeled accordingly. Note that these plots only encompass approximately half of the full longitudinal extent of the model domain.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
As in Fig. 2, but for the SLHF (color fill; every 20 W m−2) and the SSHF (green lines; every 20 W m−2). Cyclone-relative surface wind vectors appear in black arrows.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
The upstream cyclone development is initiated at the surface, which is apparent in the pressure–longitude transect time series of the meridional wind from 168 to 264 h every 48 h (Fig. 3). In contrast, the downstream cyclone is seen to initiate from the upper levels and extend down to the surface over time (Fig. 3). These cyclone evolutions are consistent with previous work (Simmons and Hoskins 1979; Orlanski and Chang 1993; Shapiro et al. 1999; Wernli et al. 1999; Wirth et al. 2018) that have shown that “in a baroclinic atmosphere a localized perturbation will disperse downstream in the upper troposphere and upstream near the surface, leading to an overall spreading of originally localized perturbations” (Wirth et al. 2018). Additionally, Fig. 3 demonstrates that the extent of the longitudinal domain is sufficiently large such that the downstream effects do not wrap around the periodic domain and impact the upstream development.
The vertically integrated water vapor (IWV) evolution in the CTRL simulation is shown in Fig. 2 for the cyclone cluster. The associated ARs with each cyclone in the cluster are clearly visible from the tongues of enhanced moisture content extending from southwest to the northeast. In an airstream perspective, the head of the AR will feed into the WCB as the warm moist air converges with and ascends over the warm front as discussed in both the WCB literature (Wernli 1997; Wernli and Davies 1997; Eckhardt et al. 2004; Schemm et al. 2013; Schemm and Wernli 2014; Binder et al. 2016; Dacre et al. 2019) and the AR literature (Ralph et al. 2011; Cordeira et al. 2013; Doyle et al. 2014, 2019; Lavers et al. 2018; Reynolds et al. 2019; Zhang et al. 2019; Zhang and Ralph 2021; Demirdjian et al. 2020, 2021). While Fig. 2 may make it appear that the subtropics are drying in terms of water vapor, this is merely an artifact resulting from continually adjusting the axis labels in Fig. 2 such that the primary cyclone appears to be at a fixed latitude and longitude. In Earth-relative space, the cyclone cluster is moving northward away from the subtropical moisture source and moves approximately 12° over the 6-day period shown in Fig. 2.
The SLHF is strongly spatially and temporally correlated with the SSHF (Fig. 4) in the CTRL simulation. The correlations are expected since both the SLHF and SSHF are both proportional to the strength of the surface wind above a region of either vapor pressure deficit or air–sea temperature contrast, respectively. The cyclone-relative surface wind vectors indicate that the air from the primary cyclone’s postfrontal sector (approximately 40°N, 175°W) flows south of the anticyclone through a region of strong (moderate) SLHF (SSHF) into the upstream cyclone (seen most clearly in Figs. 4c,d). This pathway will be referred to as the primary-to-upstream (PTU) airstream. Notably, a similar pathway can also be seen flowing from the downstream cyclone and into the primary cyclone. However, this study will concentrate only on the PTU airstream with an emphasis on its preconditioning and the resulting impact on the development of the upstream cyclone.
4. Impact of surface fluxes on upstream cyclone development
a. Control versus no surface fluxes
The impacts of the SLHF and SSHF are investigated first by comparing two simulations: (i) the CTRL (described in section 3) with full surface fluxes enabled, and (ii) the No Surface Fluxes (No SF) with both the SLHF and SSHF switched off at t = 0 h. The IWV and SLP from these two simulations are shown at t = 264 h (Fig. 5) whereby the minimum central pressure of the upstream cyclone in the CTRL case is 996 hPa, as compared to 1004 hPa for the No SF case. In contrast, the impact of the surface fluxes is substantially smaller on the primary and downstream cyclones than on the upstream cyclone, resulting in only a few hPa differences in central pressure. Despite the small differences in the primary and downstream cyclones between the two simulations, the ARs associated with each of the three cyclones are substantially weaker in the No SF simulation with max integrated vapor transport (IVT) values above 500 kg m−1 s−1 in the CTRL and around 300 kg m−1 s−1 in the No SF simulation.
Cyclone-relative plan-view maps of the IWV (color fill; every 6 mm) and SLP (black lines; every 4 hPa) for the (a) CTRL simulation at 264 h and (b) No SF simulation at 264 h. The 120-h forward trajectories are initialized at 144 h starting at the filled black circles and colored by the water vapor mixing ratio along its path (brown and blueish-green color bar). All trajectories are cyclone relative.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
To investigate the impact of the moisture source region on the upstream cyclone, forward trajectories of 120-h duration are initialized at 250-m altitude in the postfrontal sector of the primary cyclone for both simulations. The cyclone-relative trajectories colored by the water vapor mixing ratio along the trajectory are shown in Fig. 5. The PTU airstream described in section 3, through which the air from the postfrontal sector of the primary cyclone is transported south of the anticyclone and into the upstream cyclone, is illustrated in the trajectories of both simulations. It is important to note that this airstream would not be nearly as apparent if the trajectories were plotted in Earth-relative space instead of the cyclone-relative space used here. Additionally, while the trajectories appear to travel directly across the anticyclone, this is an artifact of the trajectories being overlaid onto an instantaneous field. Although the trajectories in both simulations moisten as they traverse westward (in cyclone-relative space) south of the anticyclone, the moistening along this path in the CTRL is greater than the No SF simulation by approximately 7 g kg−1. Consequently, this results in greater moisture transport to the upstream cyclone and thereby an enhanced deepening of the cyclone through stronger moist processes as will be shown in section 4d. The trajectories eventually transition into the FA whereby they bifurcate into a lower branch that continues westward retaining high moisture content, and an upper branch that turns anticyclonically toward the warm frontal convergence zone, rapidly ascends over the front, and subsequently dries (transition from blueish-green to brown colors in Fig. 5) due to precipitation. These trajectories demonstrate that the PTU airstream encompasses the FA and thereby feeds into the AR and WCB. We will also show that it is the preconditioning of the PTU airstream that serves to impact the strength of the upstream cyclone.
The trajectories from Fig. 5 are plotted in height–longitude space (Figs. 6a,b), Δqυ–longitude space (Figs. 6c,d), and ΔT–longitude space (Figs. 6e,f) in order to illustrate how the height, water vapor mixing ratio, and temperature of air parcels change as a function of longitude in each simulation. Here, Δqυ and ΔT are defined using hourly output as the change in water vapor mixing ratio and change in temperature along the trajectory from the trajectory starting positions at 144 h. In Figs. 6a and 6b the trajectories of both simulations descend from 250 m to the surface, travel west along the surface south of the anticyclone, and bifurcate whereby the upper branch rapidly ascends after connecting into the AR and WCB. The main difference between the two simulations is that the CTRL trajectories ascend a few hundred meters higher and extend a few degrees farther west than the No SF trajectories.
Diagrams along the trajectories shown in Fig. 5 for (a),(b) height, (c),(d) water vapor mixing ratio relative to trajectory initialization time, and (e),(f) temperature relative to trajectory initialization time for the (a),(c),(e) CTRL and (b),(d),(f) No SF simulations. Trajectories are 120 h long and are initialized at 144 h starting at the filled black circles.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
In the Δqυ–longitude diagrams, the CTRL trajectories (Fig. 6c) moisten by ∼4 g kg−1 early on as they descend and travel south, then continue to moisten by another ∼5 g kg−1 as they travel westward south of the anticyclone. Finally, the northern branch of trajectories flow into the AR where they connect to the WCB leading to ascent and a water vapor mixing ratio reduction of ∼7 g kg−1 through condensation. Contrast this with the No SF simulation (Fig. 6d), where the trajectories moisten by only ∼2 g kg−1 as they descend and travel south, followed by no change to the moisture as they travel westward south of the anticyclone. The trajectories in No SF finally feed into the AR where they connect to the WCB leading to ascent and a reduction in water vapor mixing ratio of only ∼3 g kg−1 to levels below their initial values through condensation. Because the trajectories in both simulations are found to moisten as they descend to the surface and travel southward (Figs. 6c,d), it is inferred that this is the result of mixing processes with surrounding air of greater moisture content since the No SF simulation has the surface fluxes switched off. Similarly, because only the CTRL simulation trajectories moisten as they travel westward and south of the anticyclone, it is apparent that the dominant reason for this moistening is due to the upward SLHF into the PTU airstream. Analogous to the moisture field, the trajectories in the CTRL undergo substantially greater warming compared with the No SF simulation as they travel westward south of the anticyclone (Figs. 6e,f). The comparison of the trajectories between the two simulations demonstrates the primary role of surface fluxes and the secondary role of mixing processes in thermodynamically preconditioning the PTU airstream. As will be discussed in section 4d, the intensity of the upstream cyclone is directly connected to this preconditioning.
We examine the impact of this moisture delivery to the AR and WCB associated with the upstream cyclone by comparing the CTRL and the No SF meridional cross sections (Fig. 7). At 192 h, the CTRL and the No SF simulations have very different moisture profiles below 800 hPa. Whereas the moist tongue in the CTRL extends far north and vertically down to the surface, the No SF simulation is much drier and has an elevated moisture maximum at ∼975 hPa (Figs. 7a,d). By 216 h, the upstream cyclone in the CTRL has begun to intensify, resulting in strong ascent at the warm front and an increase in precipitation (Fig. 7b). In contrast, the No SF simulation has only weakly intensified, resulting in little change to the warm frontal ascent and a smaller precipitation maximum (Fig. 7e). Finally at 240 h, the warm frontal ascent and precipitation maximum in the CTRL continue to increase as the upstream cyclone intensifies (Fig. 7c). Once again, the No SF simulation’s minimal cyclone intensification results in little change in the AR or WCB (Fig. 7f). This provides evidence for the hypothesis that the preconditioning of the PTU airstream and subsequent moisture transport into the upstream cyclone can substantially impact the WCB’s ascent, precipitation, and ultimately the intensity of the upstream cyclone.
Meridional cross sections along the line in Fig. 2c of water vapor mixing ratio (color fill; every 0.5 g kg−1), potential temperature (black lines; every 4 K), and vertical velocity (red lines; every 0.01 m s−1). The 24-h accumulated precipitation as a function of latitude appears at the bottom of each panel. Fields are plotted at (a),(d) 192, (b),(e) 216, and (c),(f) 240 h in the (a)–(c) CTRL and (d)–(f) No SF simulations.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
Finally, the individual impacts of the SLHF and SSHF are investigated by switching them off independently of one another. The four simulations examined here include the (i) CTRL, (ii) No SF, (iii) No SSHF, and (iv) No SLHF, which are summarized in Table 1. Figure 8 shows time series of the minimum SLP, maximum IVT, and 24-h accumulated precipitation averaged in a box (10° × 10°) centered on the upstream cyclone for each simulation. The SLHF is seen to have a greater impact on the intensity of the upstream cyclone, though the SSHF does have a nonnegligible contribution to upstream cyclone intensification (Fig. 8a). Additionally, when the SLHF is switched on we find an increase of the moisture transport and precipitation within the upstream cyclone (Figs. 8b,c). Physically, the dryness of the PTU airstream as it branches off from the primary cyclone results in a large vapor pressure deficit between the atmosphere and the sea surface, which drives strong upward SLHF into the airstream. In the No SLHF simulation, the upward SSHF south of the anticyclone warms the PTU airstream, allowing for an increased capacity of moisture content through the Clausius–Clapeyron relationship. Although warmer air parcels within the PTU airstream have the capacity to hold more moisture, we find that this ability to moisten from processes like mixing is not realized when SLHF is switched off. To demonstrate this, diagrams of along trajectory variables (initialized as in Fig. 5) are shown for the No SLHF and No SSHF simulations (Fig. 9). While both simulations are found to substantially warm (Figs. 9e,f), it is only when the SLHF is switched on that the trajectories are also found to moisten (Figs. 9c,d). Since the intensity of the upstream cyclone is modulated by moist dynamics, it is this upward SLHF into the PTU airstream that leads to a strengthening of the cyclone.
A time series of the CTRL, No SF, No SSHF, and No SLHF of the (a) minimum SLP, (b) maximum IVT, and (c) averaged 24-h precipitation in a 10° × 10° box centered on the upstream cyclone.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
As in Fig. 6, but for the (a),(c),(e) No SSHF and (b),(d),(f) No SLHF simulations.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
b. Local versus nonlocal fluxes
In the previous section, the impact of the SLHF and SSHF on the intensity of the upstream cyclone was investigated by switching the fluxes off everywhere within the domain. Given that the SLHFs are largest in the postfrontal sector of the primary cyclone and south of the high pressure center (Fig. 4), one may assume that the surface fluxes in these regions are the primary contributors to upstream cyclone intensification. However, it is possible that the surface fluxes local to the upstream cyclone may also contribute significantly to upstream cyclone intensification. To test this alternative hypothesis, the local impact region is defined as a circle of radius 8° [890 km; consistent with the 7.5° cyclone radius used in Dacre et al. (2019)] around the upstream cyclone. Everything outside of this circle is defined as the nonlocal impact region. Using this setup, a set of five simulations (described in Table 1) are considered that are the (i) CTRL (same as in sections 3 and 4a), (ii) No SF (same as in section 4a), (iii) No SF 96h, (iv) No local SF 96h, and (v) No nonlocal SF 96h. Here the “96h” refers to the time that the surface fluxes are switched off, “No local” refers to the surface fluxes being switched off inside the local impact region and on everywhere else, and “No nonlocal” refers to the surface fluxes being switched on inside the local impact region and off everywhere else. The time 96 h was chosen because it is long enough to allow for the surface fluxes to contribute toward the simulation to spin up, but without contributing significantly toward cyclone development. In Fig. 10 we show a time series of the minimum SLP, maximum IVT, and 24 h accumulated precipitation averaged in a box (10° × 10°) centered on the upstream cyclone for each of these five simulations. The simulations separate into two groups: the first group contains the CTRL and the No local SF 96h simulations, which both have the nonlocal surface fluxes switched on, and the second group contains the No SF, No SF 96h, and No nonlocal SF 96h simulations, which all have the nonlocal surface fluxes switched off. Given that the first group of simulations attain more intense upstream cyclones, greater IVT, and greater precipitation, we conclude that the surface fluxes from the nonlocal region are the primary contributors to the upstream cyclone intensity enhancement. Furthermore, the specific regions in which nonlocal fluxes are believed to be most relevant are in the postfrontal region of the primary cyclone and south of the anticyclone as discussed in sections 3 and 4a. One minor point of interest is that the No local SF 96h actually attains a slightly deeper cyclone (Fig. 10a) and accumulates a little bit more precipitation (Fig. 10c) than the CTRL run. This is attributed to the fact that the weak downward fluxes present in the upstream cyclone (Figs. 4c,d) are not present in the No local SF 96h simulation, allowing for slightly greater moisture delivery to the cyclone.
As in Fig. 8, but for the CTRL, No SF, No SF 96h, No local SF 96h, and No nonlocal SF 96h simulations.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
The sensitivity of the previous results to the circle radius of the local impact region is tested with analogous simulations having radii of 6° and 10°. This creates a set of six simulations that are the No local SF 96h with (i) r = 6°, (ii) r = 8°, (iii) r = 10°, and the No nonlocal SF 96h with (iv) r = 6°, (v) r = 8°, and (vi) r = 10°. Again, a time series is created for the minimum SLP, maximum IVT, and 24-h accumulated precipitation averaged in a box (10° × 10°) centered on the upstream cyclone (online supplemental Fig. 1). The No local SF 96h simulations exhibit no sensitivity to the radius of the local impact region. This lack of sensitivity is explained by the fact that the strongest upward fluxes of heat and moisture into the PTU airstream are outside of even the large 10° radius of the local impact region. The No nonlocal SF 96h simulations exhibit a small amount of sensitivity to the radius of the local impact region. This sensitivity exists because as the radius is increased, it simply includes more region of upward fluxes. Any additional region of upward fluxes is significant because the PTU airstream is very dry, due to the fact that nonlocal surface fluxes are switched off, as it enters the local impact region.
c. Preconditioning of the PTU airstream
The PTU airstream connecting the primary cyclone to the upstream cyclone is investigated further by initializing forward and backward trajectories at various altitudes to determine the origin of the dry air and to look more closely at the preconditioning process. First, trajectories are initialized for the CTRL simulation at 144 h and 0 m in altitude in a zonal line along 37°N from 160° to 175°W and run backward 60 h and forward 120 h (Fig. 11a). They are initialized at 144 h so that the forward trajectories arrive at the upstream cyclone during its development phase. As seen in Fig. 11a, the forward trajectories travel westward to the south of the anticyclone, moisten due to the upward surface fluxes into the PTU airstream, and eventually feed into the AR and WCB where they ascend and dry due to condensation. The backward trajectories wrap around the northwestern flank of the primary cyclone and originate north of that same cyclone. An additional set of trajectories are initialized at the same locations except starting at 1-km elevation (Fig. 11b). Unlike the previous set shown in Fig. 11a, these trajectories originate from northwest of the anticyclone and descend down to the surface where they proceed to wrap around the southern flank of the anticyclone. Strong upward surface fluxes south of the anticyclone cause the trajectories to moisten substantially, as described in section 4a, before they feed into the AR and WCB. A third set of trajectories are again initialized at the same locations except starting at 3-km elevation (not shown). However, these trajectories do not feed into the PTU airstream, but instead remain above the boundary layer and simply follow the trough–ridge–trough pattern, eventually feeding into the downstream cyclone where they ascend up above 5 km. This analysis demonstrates that the upward surface fluxes south of the anticyclone and in the postfrontal region of the primary cyclone are important contributors toward the upstream cyclone development because they act to thermodynamically precondition the PTU airstream, which was found to originate as a descending airstream from just above the boundary layer both north of the anticyclone and north of the primary cyclone.
(top) Plan view of the (a) surface wet-bulb potential temperature (color fill; every 4 K), SLP at 144 h (red lines; every 4 hPa), and SLP at 264 h (black lines; every 4 hPa) and (b) 1000-m wet-bulb potential temperature (color fill; every 4 K) and 1000-m pressure (black lines; every 4 hPa) at 264 h for the CTRL simulation. Forward and backward trajectories are colored by water vapor mixing ratio and are initialized at (a) 0- and (b) 1000-m altitude at the positions of the black filled circles. (bottom) Height–longitude diagram of the forward and backward trajectories (colored by water vapor mixing ratio) initialized at the black filled circles. All trajectories are initialized at 144 h.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
d. Dynamical impacts of moisture transport into the upstream cyclone
Thus far, we have established that the preconditioning of the PTU airstream results in enhanced moisture transport into the upstream cyclone. We have also demonstrated that the upstream cyclone is substantially weaker without the preconditioning from the surface fluxes in the cold sector of the primary cyclone and south of the surface anticyclone. However, we have not yet shown dynamically how the moisture delivery into the upstream cyclone leads to enhanced cyclone development. To show this, the same air parcel trajectories initialized in Fig. 11 are plotted at instantaneous times in a domain centered on the upstream cyclone (Fig. 12). Here, the air parcels dry substantially as they wrap cyclonically around the upstream cyclone. Simultaneously, there is an increase in both the 850-hPa diabatic heating rate and PV, both of which also wrap around the cyclone and are collocated with the air parcels. This suggests that the loss in air parcel’s water vapor content releases latent heat and drives the generation of low-level diabatic PV, which serves to enhance the strength of the upstream cyclone. Any air parcels that are within 5° latitude of the zonal dashed line in Fig. 12a are plotted in height–longitude cross sections (Fig. 13). This perspective shows that the air parcels are being lifted east of the cyclone center near the frontal occlusion zone in the region of maximum diabatic heating. The diabatic generation of low-level PV is also evident since the PV is seen to increase below the maximum in diabatic heating, where the vertical gradient in diabatic heating is largest (Hoskins et al. 1985; Lackmann 2002). Analogous plots to Figs. 12 and 13 for the No SF simulation (Figs. 14 and 15) show that the air parcels are much drier, due to the lack of SLHF, by the time they reach the upstream cyclone. As a result of the reduced water vapor content in the air parcels, there is far less diabatic heating and therefore little production of diabatic PV at the surface resulting in a much weaker cyclone (Figs. 14 and 15). We therefore conclude that the preconditioning of the air parcels in the CTRL simulation contributes toward the cyclone development through the condensation of moisture east of the cyclone center, leading to latent heating and the diabatic generation of low-level PV.
Plan-view maps centered on the upstream cyclone every 12 h of the 850-hPa PV (color fill; every 0.2 PVU; 1 PVU = 10−6 K kg−1 m2 s−1), SLP (black lines; every 4 hPa), and diabatic heating (blue lines; every 0.2 K h−1) for the CTRL simulation at (a) 210, (b) 222, (c) 234, and (d) 246 h. The air parcels for the trajectories initialized in Fig. 11 are in the filled circles colored by the water vapor mixing ratio. The location of the cross section used in Fig. 13 is shown in a dashed black line in (a).
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
Cross sections as along the dashed line shown in Fig. 12a every 12 h of the PV (color fill; every 0.2 PVU), potential temperature (black lines; every 4 K), and diabatic heating (blue lines; every 0.2 K h−1) for the CTRL simulation (a) 210, (b) 222, (c) 234, and (d) 246 h. Filled circles indicate the positions of the combined sets of trajectories from Fig. 11 that are within 5° latitude of the dashed line in Fig. 12a, and are colored according to water vapor mixing ratio.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
As in Fig. 12, but for the No SF simulation.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
As in Fig. 13, but for the No SF simulation.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
5. Discussion, summary, and conclusions
We initialize an idealized baroclinic wave in a channel using the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS) model in order to investigate the role of the surface latent heat flux (SLHF) and surface sensible heat flux (SSHF) in upstream cyclone development. Three cyclones develop within the model (Fig. 2d): the primary cyclone resulting from a prescribed localized jet stream perturbation at the initial time, the downstream cyclone formed via downstream Rossby wave propagation at the tropopause level, and the upstream cyclone formed via upstream Rossby wave propagation at the surface level. LAGRANTO trajectories are used to identify the primary-to-upstream cyclone (PTU) airstream at low levels, which we found influences the development of the upstream cyclone. A control (CTRL) simulation is compared with a No Surface Fluxes (No SF) simulation in which both SLHF and SSHF are switched off at the initial time to investigate the contribution of surface fluxes on the thermodynamic preconditioning of the PTU airstream and the intensification of the upstream cyclone. Upward surface fluxes into the PTU airstream are found to substantially moisten air parcels that are later fed directly into the atmospheric river (AR) and warm conveyor belt (WCB) associated with the upstream cyclone. As a result, eliminating surface fluxes everywhere in the model in the No SF simulation substantially reduces surface moisture content, warm frontal ascent, and precipitation relative to the CTRL simulation. Analyzing the impact of the SLHF and SSHF separately indicates that the SLHF has substantially greater impact on the upstream cyclone development than the SSHF, although the SSHF has nonnegligible contributions. The impact of local versus nonlocal surface fluxes is investigated by controlling the surface fluxes within and outside the local area around the upstream cyclone. Crucially, nonlocal surface fluxes are found to be most impactful on the upstream cyclone development. Specifically, surface fluxes to the south of the anticyclone and in the postfrontal sector of the primary cyclone are shown to be of greatest importance for the intensity of the upstream cyclone. Backward and forward trajectories show that the PTU air parcels originate from a broad region north of the primary cyclone and anticyclone. Finally, the preconditioning of the PTU airstream is found to lead to increased moisture delivery to the upstream cyclone. This results in increased condensational heating followed by increased diabatic generation of PV, which serves to further intensify the upstream cyclone.
A schematic representation that summarizes the key physical processes associated with the PTU airstream is shown in Fig. 16. The PTU airstream results from the confluence of two dry airstreams. The first originates above the boundary layer north of the anticyclone and descends in the postfrontal sector of the primary cyclone. The second originates north of the primary cyclone and also travels into the postfrontal sector of the primary cyclone. As these airstreams merge, they begin to moisten due primarily to upward SLHF and, to a lesser extent, from mixing with surrounding air. The PTU airstream flows westward (in a cyclone-relative sense) within the boundary layer along the southern flank of the anticyclone while continuing to moisten from the upward SLHF. Finally, the PTU airstream feeds into the AR which connects into the WCB where air parcels ascend at the warm-frontal boundary and undergo a substantial loss of water vapor content due to condensation. The PTU is defined here as encompassing the portion of the near-surface airstream starting from the cold sector of the primary cyclone, passing along the southern flank of the anticyclone, and ending by connecting into the AR and WCB through the FA. The PTU airstream may be seen as an extension of the feeder airstream (FA) discussed in Dacre et al. (2019) for the particular case when there is a cluster of cyclones. In solitary cyclones, however, the FA occurs without the PTU extension.
Schematic representation of the relevant airstreams discussed in this study. Airstreams are illustrated as colored bands where brown is dry and blue is moist. The IWV is shown on the surface in pale colors from orange to red for small and large values, respectively, and “Atm River” is the abbreviation for atmospheric river. The surface fluxes are shown by small pink squiggly arrows, mixing processes by small black circular arrows, cold fronts by bold blue lines, and warm fronts by bold red lines. The high and low pressure centers are denoted by “H” and “L,” respectively, and sea level pressure contours are indicated by black lines. The upper-level jet stream is illustrated as an approximate single pressure contour visualizing the trough and ridges. Heights are shown on the left wall.
Citation: Journal of the Atmospheric Sciences 80, 6; 10.1175/JAS-D-22-0251.1
This study builds on the work of Boutle et al. (2010) and Papritz et al. (2021), who found that the moisture sourced in the postfrontal and anticyclonic boundary layer is transported long distances to an upstream cyclone. Whereas Boutle et al. (2010) was focused on moisture ventilation from the boundary layer into the free troposphere, Papritz et al. (2021) investigated the impacts of these moisture sources on upstream cyclone precipitation. Our analysis is unique in that it focuses on the impact of this long range moisture transport on the moist dynamics of the upstream cyclone. Furthermore, our trajectories trace the air masses farther backward than Boutle et al. (2010) or Papritz et al. (2021) to show that the dry air in the cold sector originates from both a descending air mass and from north of the primary cyclone.
While the present study has focused on the role of long range moisture transport in developing an upstream cyclone and cyclone clustering, there are other possible mechanisms at play here as well. Pinto et al. (2014) approached the problem of cyclone clusters from a jet dynamics perspective. They found that for cyclone clusters the primary cyclone tends to develop at the left jet exit region, where upper-level divergence tends to be maximized. In contrast, the secondary (upstream) cyclone forms along the primary cyclone’s trailing cold front and at the right jet entrance region of the jet streak. Priestley et al. (2017) builds on the jet dynamics perspective finding that cyclone clusters are associated with an extended and anomalously strong eddy-driven jet due to Rossby wave breaking enhancing the upper tropospheric thermal gradient and also extending the jet eastward. The approach of Weijenborg and Spengler (2020) is more in line with our own whereby they propose that cyclone clustering is a result of diabatic generation of baroclinicity. Our results differ from Weijenborg and Spengler (2020) in that while they were focused on demonstrating how latent heating can serve to increase the baroclinicity of the storm tracks, we directly connect the moisture source regions to the intensification of the upstream cyclone. It is important to note that the upper-level jet and moist dynamical perspectives for cyclone clustering are not mutually exclusive and the two concepts can even coexist and potentially reinforce one another.
We hypothesize that the dry air intrusion (DI), defined by an airstream that descends ≥400 hPa in 48 h (Raveh-Rubin 2017; Raveh-Rubin and Catto 2019; Catto and Raveh-Rubin 2019; Ilotoviz et al. 2021), can connect into the PTU airstream since it is known to also descend into the postfrontal region of an extratropical cyclone (Browning 1993), just like the descending airstream identified in this study (Figs. 11b and 16). If this hypothesis is correct, then it would imply that the preconditioning of the DI may be as important for upstream cyclone intensification as the preconditioning of the dry airstreams originating at lower altitudes above the boundary layer identified in this study.
It is important to keep the limitations of idealized configurations in mind when interpreting the result of this study. First and foremost, a cluster of cyclones in the real atmosphere is never as isolated and simple as occurs in an idealized baroclinic channel. As a result, the tendency for the model framework employed in this study to produce clusters of cyclones and anticyclones through downstream and upstream Rossby wave energy dispersion, and with clearly defined airstreams connecting them, may be greater than in the real atmosphere. The upstream cyclone development in particular, which is due to the upstream propagation of surface Rossby waves, is not frequently or easily observed in the real atmosphere (Wirth et al. 2018). Nevertheless, this highly controlled framework enables us to isolate key process involved in upstream cyclogenesis, which is more poorly understood than its downstream counterpart. While surface Rossby waves did lead to the formation of the upstream cyclone investigated here, there are other mechanisms for upstream cyclone formation and the results of this study should be relevant for any pair of cyclones in close proximity to one another. Another limitation of our modeling framework is that the SST is set to be constant and zonally symmetric. Zonal asymmetries exist, such as western boundary currents, and are important for maintaining low-level baroclinicity that promotes the formation of cyclone clusters even in the absence of a zonally continuous upper-level jet. The SST is also set to be constant in time in this study, which eliminates the coupled air–sea processes that influence surface fluxes. Finally, the model does not include radiation processes, which prevents these simulations from representing the diurnal heating or cooling associated with clouds and the free atmosphere that results in a diurnal cycle of convection. Despite these limitations, the idealized framework provides the controlled environment needed to isolate the key characteristics of the PTU airstream and its impact on upstream cyclones and to build foundational understanding that can be applied in more realistic case studies of upstream cyclogenesis events.
Acknowledgments.
We thank Daniel and Aly Demirdjian for their art direction and advice on illustrating the schematic. Additionally, we thank Jason M. Cordeira for providing insightful tips that helped to improve the accuracy of our schematic. We also would like to thank Dale Durran and Daniel Lloveras for their help with setting up the initial-condition jet configuration. This research was performed while the author held an NRC Research Associateship award at the Naval Research Laboratory in Monterey, California. We gratefully acknowledge the support of the NRL Base Program, PE 0601153N. Computational resources were provided by the Navy Department of Defense Supercomputing Resource Center in Stennis, Mississippi.
Data availability statement.
Due to confidentiality agreements, supporting data can only be made available to bona fide researchers subject to a nondisclosure agreement. Details of the data and how to request access are available from Reuben Demirdjian at the U.S. Naval Research Laboratory.
REFERENCES
Binder, H., M. Boettcher, H. Joos, and H. Wernli, 2016: The role of warm conveyor belts for the intensification of extratropical cyclones in Northern Hemisphere winter. J. Atmos. Sci., 73, 3997–4020, https://doi.org/10.1175/JAS-D-15-0302.1.
Bjerknes, J., and H. Solberg, 1922: Life cycle of cyclones and the polar front theory of atmospheric circulation. Geofys. Publ., 3, 468–473.
Booth, J. F., L. Thompson, J. Patoux, and K. A. Kelly, 2012: Sensitivity of midlatitude storm intensification to perturbations in the sea surface temperature near the Gulf Stream. Mon. Wea. Rev., 140, 1241–1256, https://doi.org/10.1175/MWR-D-11-00195.1.
Boutle, I. A., R. J. Beare, S. E. Belcher, A. R. Brown, and R. S. Plant, 2010: The moist boundary layer under a mid-latitude weather system. Bound.-Layer Meteor., 134, 367–386, https://doi.org/10.1007/s10546-009-9452-9.
Browning, K. A., 1993: The dry intrusion perspective of extra-tropical cyclone development. Meteor. Appl., 4, 317–324, https://doi.org/10.1017/S1350482797000613.
Bui, H., and T. Spengler, 2021: On the influence of sea surface temperature distributions on the development of extratropical cyclones. J. Atmos. Sci., 78, 1173–1188, https://doi.org/10.1175/JAS-D-20-0137.1.
Carrera, M. L., J. R. Gyakum, and D. Zhang, 1999: A numerical case study of secondary marine cyclogenesis sensitivity to initial error and varying physical processes. Mon. Wea. Rev., 127, 641–660, https://doi.org/10.1175/1520-0493(1999)127<0641:ANCSOS>2.0.CO;2.
Catto, J. L., and S. Raveh-Rubin, 2019: Climatology and dynamics of the link between dry intrusions and cold fronts during winter. Part I: Global climatology. Climate Dyn., 53, 1873–1892, https://doi.org/10.1007/s00382-019-04745-w.
Cordeira, J. M., F. M. Ralph, and B. J. Moore, 2013: The development and evolution of two atmospheric rivers in proximity to western North Pacific tropical cyclones in October 2010. Mon. Wea. Rev., 141, 4234–4255, https://doi.org/10.1175/MWR-D-13-00019.1.
Dacre, H. F., and J. G. Pinto, 2020: Serial clustering of extratropical cyclones: A review of where, when and why it occurs. npj Climate Atmos. Sci., 3, 48, https://doi.org/10.1038/s41612-020-00152-9.
Dacre, H. F., O. Martínez-Alvarado, and C. O. Mbengue, 2019: Linking atmospheric rivers and warm conveyor belt airflows. J. Hydrometeor., 20, 1183–1196, https://doi.org/10.1175/JHM-D-18-0175.1.
Davies, H. C., C. Schär, and H. Wernli, 1991: The palette of fronts and cyclones within a baroclinic wave development. J. Atmos. Sci., 48, 1666–1689, https://doi.org/10.1175/1520-0469(1991)048<1666:TPOFAC>2.0.CO;2.
Demirdjian, R., J. D. Doyle, C. A. Reynolds, J. R. Norris, A. C. Michaelis, and F. M. Ralph, 2020: A case study of the physical processes associated with the atmospheric river initial-condition sensitivity from an adjoint model. J. Atmos. Sci., 77, 691–709, https://doi.org/10.1175/JAS-D-19-0155.1.
Demirdjian, R., R. Rotunno, B. D. Cornuelle, C. A. Reynolds, and J. D. Doyle, 2021: The circulation response of a two-dimensional frontogenetic model to optimized moisture perturbations. J. Atmos. Sci., 78, 459–472, https://doi.org/10.1175/JAS-D-20-0102.1.
Demirdjian, R., J. D. Doyle, P. M. Finocchio, and C. A. Reynolds, 2022: On the influence of surface latent heat fluxes on idealized extratropical cyclones. J. Atmos. Sci., 79, 2229–2242, https://doi.org/10.1175/JAS-D-22-0035.1.
Doyle, J. D., C. Amerault, C. A. Reynolds, and P. A. Reinecke, 2014: Initial condition sensitivity and predictability of a severe extratropical cyclone using a moist adjoint. Mon. Wea. Rev., 142, 320–342, https://doi.org/10.1175/MWR-D-13-00201.1.
Doyle, J. D., C. A. Reynolds, and C. Amerault, 2019: Adjoint sensitivity analysis of high-impact extratropical cyclones. Mon. Wea. Rev., 147, 4511–4532, https://doi.org/10.1175/MWR-D-19-0055.1.
Eckhardt, S., A. Stohl, H. Wernli, P. James, C. Forster, and N. Spichtinger, 2004: A 15-year climatology of warm conveyor belts. J. Climate, 17, 218–237, https://doi.org/10.1175/1520-0442(2004)017<0218:AYCOWC>2.0.CO;2.
Finocchio, P. M., and J. D. Doyle, 2019: How the speed and latitude of the jet stream affect the downstream response to recurving tropical cyclones. Mon. Wea. Rev., 147, 3261–3281, https://doi.org/10.1175/MWR-D-19-0049.1.
Fish, M. A., A. M. Wilson, and F. M. Ralph, 2019: Atmospheric river families: Definition and associated synoptic conditions. J. Hydrometeor., 20, 2091–2108, https://doi.org/10.1175/JHM-D-18-0217.1.
Gyakum, J. R., and R. E. Danielson, 2000: Analysis of meteorological precursors to ordinary and explosive cyclogenesis in the western North Pacific. Mon. Wea. Rev., 128, 851–863, https://doi.org/10.1175/1520-0493(2000)128<0851:AOMPTO>2.0.CO;2.
Hirata, H., R. Kawamura, M. Kato, and T. Shinoda, 2015: Influential role of moisture supply from the Kuroshio/Kuroshio Extension in the rapid development of an extratropical cyclone. Mon. Wea. Rev., 143, 4126–4144, https://doi.org/10.1175/MWR-D-15-0016.1.
Hodur, R. M., 1997: The Naval Research Laboratory’s Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS). Mon. Wea. Rev., 125, 1414–1430, https://doi.org/10.1175/1520-0493(1997)125<1414:TNRLSC>2.0.CO;2.
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, 877–946, https://doi.org/10.1002/qj.49711147002.
Ilotoviz, E., V. P. Ghate, and S. Raveh‐Rubin, 2021: The impact of slantwise descending dry intrusions on the marine boundary layer and air‐sea interface over the ARM eastern North Atlantic site. J. Geophys. Res. Atmos., 126, e2020JD033879, https://doi.org/10.1029/2020JD033879.
Kuo, Y.-H., S. Low-Nam, and R. J. Reed, 1991: Effects of surface energy fluxes during the early development and rapid intensification stages of seven explosive cyclones in the western Atlantic. Mon. Wea. Rev., 119, 457–476, https://doi.org/10.1175/1520-0493(1991)119<0457:EOSEFD>2.0.CO;2.
Lackmann, G. M., 2002: Cold-frontal potential vorticity maxima, the low-level jet, and moisture transport in extratropical cyclones. Mon. Wea. Rev., 130, 59–74, https://doi.org/10.1175/1520-0493(2002)130<0059:CFPVMT>2.0.CO;2.
Lavers, D. A., M. J. Rodwell, D. S. Richardson, F. M. Ralph, J. D. Doyle, C. A. Reynolds, V. Tallapragada, and F. Pappenberger, 2018: The gauging and modeling of rivers in the sky. Geophys. Res. Lett., 45, 7828–7834, https://doi.org/10.1029/2018GL079019.
Lloveras, D. J., L. H. Tierney, and D. R. Durran, 2022: Mesoscale predictability in moist midlatitude cyclones is not sensitive to the slope of the background kinetic energy spectrum. J. Atmos. Sci., 79, 119–139, https://doi.org/10.1175/JAS-D-21-0147.1.
Manola, I., F. Selten, H. De Vries, and W. Hazeleger, 2013: “Waveguidability” of idealized jets. J. Geophys. Res. Atmos., 118, 10 432–10 440, https://doi.org/10.1002/jgrd.50758.
Neiman, P. J., M. A. Shapiro, E. G. Donall, and C. W. Kreitzberg, 1990: Diabatic modification of an extratropical marine cyclone warm sector by cold underlying water. Mon. Wea. Rev., 118, 1576–1590, https://doi.org/10.1175/1520-0493(1990)118<1576:DMOAEM>2.0.CO;2.
Nuss, W. A., 1989: Air–sea interaction influences on the structure and intensification of an idealized marine cyclone. Mon. Wea. Rev., 117, 351–369, https://doi.org/10.1175/1520-0493(1989)117<0351:ASIIOT>2.0.CO;2.
Orlanski, I., and E. K. M. Chang, 1993: Ageostrophic geopotential fluxes in downstream and upstream development of baroclinic waves. J. Atmos. Sci., 50, 212–225, https://doi.org/10.1175/1520-0469(1993)050<0212:AGFIDA>2.0.CO;2.
Papritz, L., F. Aemisegger, and H. Wernli, 2021: Sources and transport pathways of precipitating waters in cold-season deep North Atlantic cyclones. J. Atmos. Sci., 78, 3349–3368, https://doi.org/10.1175/JAS-D-21-0105.1.
Phillips, B., and L. O’Neill, 2020: Observational analysis of extratropical cyclone interactions with northeast Pacific sea surface temperature anomalies. J. Climate, 33, 6745–6763, https://doi.org/10.1175/JCLI-D-19-0853.1.
Pinto, J. G., I. Gómara, G. Masato, H. F. Dacre, T. Woollings, and R. Caballero, 2014: Large‐scale dynamics associated with clustering of extratropical cyclones affecting western Europe. J. Geophys. Res. Atmos., 119, 13 704–13 719, https://doi.org/10.1002/2014JD022305.
Polvani, L. M., and J. G. Esler, 2007: Transport and mixing of chemical air masses in idealized baroclinic life cycles. J. Geophys. Res., 112, D23102, https://doi.org/10.1029/2007JD008555.
Priestley, M. D., J. G. Pinto, H. F. Dacre, and L. C. Shaffrey, 2017: Rossby wave breaking, the upper level jet, and serial clustering of extratropical cyclones in western Europe. Geophys. Res. Lett., 44, 514–521, https://doi.org/10.1002/2016GL071277.
Ralph, F. M., P. J. Neiman, G. N. Kiladis, K. Weickmann, and D. W. Reynolds, 2011: A multiscale observational case study of a Pacific atmospheric river exhibiting tropical–extratropical connections and a mesoscale frontal wave. Mon. Wea. Rev., 139, 1169–1189, https://doi.org/10.1175/2010MWR3596.1.
Raveh-Rubin, S., 2017: Dry intrusions: Lagrangian climatology and dynamical impact on the planetary boundary layer. J. Climate, 30, 6661–6682, https://doi.org/10.1175/JCLI-D-16-0782.1.
Raveh-Rubin, S., and J. L. Catto, 2019: Climatology and dynamics of the link between dry intrusions and cold fronts during winter, part II: Front-centred perspective. Climate Dyn., 53, 1893–1909, https://doi.org/10.1007/s00382-019-04793-2.
Reynolds, C. A., J. D. Doyle, F. M. Ralph, and R. Demirdjian, 2019: Adjoint sensitivity of North Pacific atmospheric river forecasts. Mon. Wea. Rev., 147, 1871–1897, https://doi.org/10.1175/MWR-D-18-0347.1.
Rotunno, R., W. C. Skamarock, and C. Snyder, 1994: An analysis of frontogenesis in numerical simulations of baroclinic waves. J. Atmos. Sci., 51, 3373–3398, https://doi.org/10.1175/1520-0469(1994)051<3373:AAOFIN>2.0.CO;2.
Schemm, S., and H. Wernli, 2014: The linkage between the warm and the cold conveyor belts in an idealized extratropical cyclone. J. Atmos. Sci., 71, 1443–1459, https://doi.org/10.1175/JAS-D-13-0177.1.
Schemm, S., H. Wernli, and L. Papritz, 2013: Warm conveyor belts in idealized moist baroclinic wave simulations. J. Atmos. Sci., 70, 627–652, https://doi.org/10.1175/JAS-D-12-0147.1.
Shapiro, M. A., and Coauthors, 1999: A planetary-scale to mesoscale perspective of the life cycles of extratropical cyclones: The bridge between theory and observations. The Life Cycles of Extratropical Cyclones, M. A. Shapiro and S. Grønas, Eds., Amer. Meteor. Soc., 139–185.
Simmons, A. J., and B. J. Hoskins, 1979: The downstream and upstream development of unstable baroclinic waves. J. Atmos. Sci., 36, 1239–1254, https://doi.org/10.1175/1520-0469(1979)036<1239:TDAUDO>2.0.CO;2.
Sodemann, H., and A. Stohl, 2013: Moisture origin and meridional transport in atmospheric rivers and their association with multiple cyclones. Mon. Wea. Rev., 141, 2850–2868, https://doi.org/10.1175/MWR-D-12-00256.1.
Sprenger, M., and H. Wernli, 2015: The LAGRANTO Lagrangian analysis tool—Version 2.0. Geosci. Model Dev., 8, 2569–2586, https://doi.org/10.5194/gmd-8-2569-2015.
Thorncroft, C. D., B. J. Hoskins, and M. E. McIntyre, 1993: Two paradigms of baroclinic-wave life-cycle behaviour. Quart. J. Roy. Meteor. Soc., 119, 17–55, https://doi.org/10.1002/qj.49711950903.
Weijenborg, C., and T. Spengler, 2020: Diabatic heating as a pathway for cyclone clustering encompassing the extreme storm Dagmar. Geophys. Res. Lett., 47, e2019GL085777, https://doi.org/10.1029/2019GL085777.
Wernli, H., 1997: A Lagrangian-based analysis of extratropical cyclones. II: A detailed case-study. Quart. J. Roy. Meteor. Soc., 123, 1677–1706, https://doi.org/10.1002/qj.49712354211.
Wernli, H., and H. C. Davies, 1997: A Lagrangian-based analysis of extratropical cyclones. I: The method and some applications. Quart. J. Roy. Meteor. Soc., 123, 467–489, https://doi.org/10.1002/qj.49712353811.
Wernli, H., and C. Schwierz, 2006: Surface cyclones in the ERA-40 dataset (1958–2001). Part I: Novel identification method and global climatology. J. Atmos. Sci., 63, 2486–2507, https://doi.org/10.1175/JAS3766.1.
Wernli, H., M. A. Shapiro, and J. Schmidli, 1999: Upstream development in idealized baroclinic wave experiments. Tellus, 51A, 574–587, https://doi.org/10.3402/tellusa.v51i5.14476.
Wirth, V., M. Riemer, E. K. M. Chang, and O. Martius, 2018: Rossby wave packets on the midlatitude waveguide—A review. Mon. Wea. Rev., 146, 1965–2001, https://doi.org/10.1175/MWR-D-16-0483.1.
Xu, H., M. Xu, S. Xie, and Y. Wang, 2011: Deep atmospheric response to the spring Kuroshio over the East China Sea. J. Climate, 24, 4959–4972, https://doi.org/10.1175/JCLI-D-10-05034.1.
Zhang, Z., and F. M. Ralph, 2021: The influence of antecedent atmospheric river conditions on extratropical cyclogenesis. Mon. Wea. Rev., 149, 1337–1357, https://doi.org/10.1175/MWR-D-20-0212.1.
Zhang, Z., F. M. Ralph, and M. Zheng, 2019: The relationship between extratropical cyclone strength and atmospheric river intensity and position. Geophys. Res. Lett., 46, 1814–1823, https://doi.org/10.1029/2018GL079071.
