1. Introduction
The synoptic-scale genesis, evolution, and dynamics of extratropical cyclones (ETCs) have been well understood for over a century (e.g., Bjerknes 1919; Bjerknes and Solberg 1922; Shapiro and Keyser 1990). However, unresolved questions remain regarding how surface processes, and the environment within and around ETCs, relate to their development, intensity, and variations in the distribution of clouds and precipitation at smaller scales. ETCs can efficiently transport heat and energy from the tropical regions toward the middle and high latitudes, serving as a vital source of freshwater for these regions (Heideman and Fritsch 1988; Hawcroft et al. 2012). Additionally, ETCs are associated with nearly 90% of precipitation in storm-track regions (Catto et al. 2012) and close to 80% of extreme precipitation events (Pfahl and Wernli 2012). Given their important contribution to Earth’s climate, it is vital to continue to improve our understanding of ETCs at finer scales and from different points of view.
Latent heat flux (LHF) and sensible heat flux (SHF) over the ocean and land surfaces affect various weather and climate patterns, and they have been hypothesized to play a large role in the development of ETCs, especially those that form over the oceans near landmasses with greater air–sea thermodynamic differences. Extreme marine cyclogenesis and “bomb” cyclones typically occur downwind of cold continental landmasses near large sea surface temperature (SST) gradients (Sanders and Gyakum 1980). For example, extreme cyclogenesis often occurs over the western Atlantic and western Pacific Oceans in the Northern Hemisphere, which include the Gulf Stream and Kuroshio flowing northward parallel to North America and eastern Asia, respectively. As LHF and SHF are driven by surface winds and air–sea differences in humidity and temperature, respectively, large LHF and SHF values are typically observed in these regions of the globe (Liu et al. 1979; Yu and Weller 2007), which in turn maintain the strong near-surface baroclinicity often observed in these regions (Hotta and Nakamura 2011). The cold conveyor belt of the ETC can perform a significant role in the strengthening of LHF and SHF, as it draws in colder and dryer air from the polar region equatorward, increasing the air–sea differences and surface winds (Hirata et al. 2018). The combined surface heat fluxes induce turbulence near the surface, causing an exchange of energy, momentum, and heat between the sea surface and the lower boundary layer. As ETCs form, LHF and SHF within and around the cyclone increase the baroclinicity and instability within the boundary layer. This correlates with rapid intensification (Vukovich et al. 1991; Cione et al. 1993; Catto 2016), facilitating further latent heat release and augmenting ETC evolution (Booth et al. 2012; de Vries et al. 2019; Hirata et al. 2019).
Early field campaigns noted the correlation between combined surface heat fluxes and rapid marine cyclogenesis (Neiman and Shapiro 1993), as combined LHF and SHF exceeding 1000 W m−2 near the cyclone center reduced the static stability in the region (Vukovich et al. 1991). Modeling studies around this time attempted to address the direct impact of surface fluxes on ETC development, but led to differing results, with some finding heat fluxes could negatively affect ETC strengthening (Branscome et al. 1989) while others found that LHF and SHF may aid in their maturation (Mak 1998). Modeling studies within the past decade have indicated that combined surface heat fluxes may play a dominant role within the warm sector of ETCs by regulating the moisture and heat in the source region for the warm conveyor belt (WCB; Booth et al. 2012). These fluxes not only affect the development of the ETC near the surface but also correlate to alterations at higher vertical levels within the cyclone (Small et al. 2014). As these fluxes intensify their respective ETC, it can lead to a positive feedback loop as the cyclone’s stronger winds increase the fluxes (Hirata et al. 2015, 2018). However, while these studies have been able to shed light on potential mechanisms through which surface processes and extratropical cyclogenesis may be related, their scope has been limited due to the relative lack of observations of LHF and SHF within ETCs necessary to validate their analysis (Booth et al. 2012).
Spatially widespread and temporally frequent observations of ocean SHF and LHF are critical for our understanding of marine-based ETCs, but current flux measurements have been limited to a handful of buoys and occasional field campaigns. While these in situ observations continue to be the standard and offer detailed estimates at their representative times and locations, they do not offer sufficient spatial and temporal coverage to properly examine large-scale processes. It is also imperative to obtain finer scale observations of LHF and SHF, as they can vary on short temporal and spatial scales; fluxes, and instability associated with them, can magnify in the presence of strong SST gradients (Cione et al. 1993). Remote sensing instruments can aid in filling in these gaps where in situ data may be limited. While current spaceborne instruments cannot directly measure SHF and LHF, they can measure the components (temperature, humidity, wind speed) that are necessary to estimate the fluxes using the bulk aerodynamic formulas (Bentamy et al. 2003). Unfortunately, these instruments, such as microwave radiometers, have their limitations. Most instruments are in a polar orbit and have large spatial and temporal gaps in the tropics and lower midlatitudes, decreasing the sampling frequency over these parts of the globe. Additionally, their signals can be attenuated by strong precipitation, possibly leading to inaccurate or missing surface observations (Mears et al. 2000; Portabella et al. 2012). The recently launched the Cyclone Global Navigation Satellite System (CYGNSS) mission addresses both of these limitations and can bridge the gaps in our understanding of the surface processes over the tropical and subtropical oceans that may be present from current in situ and spaceborne instruments.
CYGNSS utilizes global positioning system (GPS) signals reflecting off the ocean surface, which do not experience significant attenuation in the presence of precipitation, to estimate surface wind speeds. As a constellation of eight small satellites in a tropical orbit (35° orbit inclination), it can provide improved surface processes coverage over the tropical and subtropical oceans (Ruf et al. 2016). Although CYGNSS is a tropical mission, previous research applying prelaunch observation statistics demonstrated that the constellation would have the ability to observe ETCs and their fronts that developed in the lower midlatitudes (~30°–40°), especially over the western boundary currents in the western Atlantic and Pacific Oceans (Crespo et al. 2017). Since its launch in December 2016, the CYGNSS Surface Heat Flux Product has been developed for the mission. This product utilizes CYGNSS’s surface wind speed observations and thermodynamic quantities from reanalysis data to estimate LHF and SHF along CYGNSS’s orbit (Crespo et al. 2019). While this product utilizes a reanalysis dataset for the thermodynamic variables needed, the numerous wind speed observations CYGNSS can provide between grid points from a typical reanalysis dataset can provide greater detail in variations of LHF and SHF.
This paper will employ both the wind speed and surface flux retrievals from CYGNSS to examine various low-latitude extratropical cyclones that it has observed since launch. While it was not designed as a midlatitude mission, this paper and its results will demonstrate CYGNSS’s ability to procure surface observations throughout the genesis and evolution of many low-latitude ETCs and will provide insight into the distribution of ocean LHF and SHF in and around these storms.
The remainder of this paper is organized as follows: section 2 features an overview of the CYGNSS mission, along with its wind speed and surface heat flux products, and describes the methods used in the composite analysis. Section 3 features three ETC case studies in different ocean basins observed by CYGNSS since launch, and section 4 focuses on a composite analysis of all ETCs observed by CYGNSS from 2017 to 2019. The summary and conclusions are contained in section 5.
2. Data and methods
a. CYGNSS
CYGNSS was launched in December 2016 and began routinely taking science data in the latter half of March 2017. CYGNSS consists of a constellation of eight small satellites in a 35° orbital inclination; it is designed to estimate surface winds over the tropical and subtropical oceans employing Global Navigation Satellite System Reflectometry (GNSS-R) technology. Every observatory measures the direct signal from the existing GPS satellite via a zenith antenna while measuring the reflected GPS signal from the ocean surface as specular points through its two (port and starboard) downward-facing antennas (Ruf et al. 2016). Each CYGNSS satellite can observe up to four specular points per second, resulting in up to 32 simultaneous observations per second across the entire constellation. The frequency of these observations changed to four specular points every half second in July 2019, resulting in up to 64 specular point observations per second for the remainder of the mission.
The scattering map created from the reflected GPS signal, with coordinated code chip delay and Doppler shift, is referred to as the delay Doppler map (DDM; Clarizia et al. 2009). The average reflected power near the specular point [the DDM average (DDMA)] and the slope of the DDM waveform in time-delayed coordinates [leading edge slope (LES)] are separately used to estimate the surface wind speeds using geophysical model functions (GMFs) (Clarizia et al. 2014; Ruf et al. 2016). The LES and DDMA winds are then optimally combined using a minimum variance (MV) estimate to produce the best estimate wind speed product (Ruf et al. 2016; Clarizia et al. 2018). Time averaging is applied to consecutive half-second frequency DDMs to smooth the data and reduce noise between specular points, with the number of samples used for the time-averaging dependent on the incidence angle of the specular point such that the effective resolution is always ~25 km.
The CYGNSS wind speed measurements assume that the sea state is in equilibrium with the wind speed and is formally referred to as the “fully developed seas” (FDS) wind speed product (CYGNSS 2020b). While the wind speed estimates are reliable throughout most of the tropical and subtropical oceans, they are often inaccurate at higher wind speeds and in rapidly developing systems (Ruf et al. 2019a,b). Therefore, an alternative wind speed product, “young seas with limited fetch” (YSLF), was introduced that does not assume that the sea state is in equilibrium with the local wind. While the YSLF wind speed product may be beneficial for the analysis of very strong (e.g., surface wind speeds of >20 m s−1) ETCs, at the time of writing the YSLF product had not yet been sufficiently tested in and around ETCs. Furthermore, as will be discussed in the following paragraph, LHF and SHF estimates are not valid at wind speeds greater than 25 m s−1.
b. CYGNSS Ocean Surface Heat Flux Product
The CYGNSS Ocean Surface Heat Flux Product (hereinafter CYGNSS Fluxes) was developed by deriving the LHF and SHF from CYGNSS’s wind speed observations, along with thermodynamic inputs (i.e., humidity and temperature) from the second version of the Modern-Era Retrospective Analysis for Research and Applications (MERRA-2) (Gelaro et al. 2017; Crespo et al. 2019; CYGNSS 2020a). These inputs from CYGNSS and MERRA-2 are then used to compute LHF and SHF via version 3.5 of the Coupled Ocean–Atmosphere Response Experiment (COARE 3.5) flux algorithm, which utilizes the bulk aerodynamic formula and parameterizes the drag coefficient as a function of wind speed (Edson et al. 2013). Through the bulk aerodynamic formulas, LHF and SHF are estimated as functions of the surface wind speed and air–sea differences in specific humidity and temperature. Due to limitations in the COARE algorithm, the CYGNSS Fluxes can only be validated for wind speeds up to 25 m s−1. This is due to it not being able to account for the impact of sea spray on fluxes at higher wind speeds (Richter and Stern 2014), and the breakdown of the drag coefficient estimate (Edson et al. 2013), which could lead to inaccurate LHF and SHF estimates. CYGNSS Flux estimates of ocean SHF and LHF have been independently validated with independent buoy observations; wind, temperature, and humidity observations from these buoys were used to estimate LHF and SHF using the same algorithm used within the CYGNSS Fluxes product. While the CYGNSS Fluxes uses both CYGNSS winds and reanalysis data, it does not hinder their performance as the results from this product validate well to the independent buoy data. Additionally, an uncertainty parameter is provided for the CYGNSS Flux product (used in section 4) by factoring in the reported uncertainties from the CYGNSS wind speed product and MERRA-2 (Crespo et al. 2019).
c. Composite analysis methods
Over the past decade, a database of ETC locations and associated fronts was developed and is used here. The database was initiated by Bauer and Del Genio (2006) who developed a cyclone tracking algorithm that they applied to the ERA-Interim reanalysis (ERA5 after 2018). The algorithm searches for local minima in sea level pressure and additionally can track the ETCs in time. A full description and evaluation can be found in Bauer et al. (2016).
In parallel, we developed a routine to identify cold and warm fronts to supplement the cyclone-location database (Naud et al. 2010, 2016). We use two published methods to identify the fronts. The first one was introduced by Hewson (1998) and isolates regions of local maximum in thermal gradient to identify both cold and warm fronts. We apply this method to the MERRA-2 reanalysis temperature fields at 850 hPa. At the onset of the front detection project, MERRA-2 was chosen instead of ERA-Interim because it provides a higher spatial resolution that is more appropriate for the finer structure of fronts. We do, however, apply a 4–1–1–1 weighted smoothing to the raw MERRA-2 output to help avoid some spurious effects from sharp gradients in temperature. Because the Hewson method experiences issues in regions of low baroclinicity (Schemm et al. 2015), we supplement detection of cold fronts by using a method proposed by Simmonds et al. (2012) who focus on a change in the surface and 850-hPa wind direction across cold fronts. A full description of the method is available in Naud et al. (2016). The full global database of 6-hourly cyclone and corresponding front locations covers the period 2006–19 (kept up to date in 1-yr increments) and is publicly accessible as described in the data availability statement (see below). We presently use a subset of this database for the years 2017–19.
In this study, we match CYGNSS observations in time and space with the 6-hourly cyclone identifications: observations must occur within ±3 h of the cyclone/front detections and at a predefined maximum distance of 500 km from the storm center and 25 km from either warm or cold fronts. When creating frontal composite transects, we use all observations taken within a Euclidian distance of ±1500 km of a given front. When creating cyclone-centered composites, we use all observations within a 1500-km radius from the storm center. In addition, this analysis only selects ETCs with a center occurring within ±40° latitude.
3. Case study analysis
We analyzed three extratropical cyclones that occurred in separate ocean basins within CYGNSS’s orbit and field of view. This includes the western Pacific and western Atlantic Oceans in the Northern Hemisphere, as well as one storm that occurred in the southern Pacific near the Australian east coast. These case studies were selected because they each featured multiple CYGNSS overpasses, providing information on the time evolution of the surface processes occurring within and around these ETCs. Additionally, the western Atlantic case was one of the strongest rapidly intensifying extratropical cyclones so far this century (Hirata et al. 2019).
a. Western Pacific Ocean case
One of the first weather systems observed by CYGNSS after its science mission commenced was an ETC located to the east of Japan around 21 March 2017 (Fig. 1). Three CYGNSS overpasses of the ETC highlight its development (around 0000 UTC 21 March, between 2100 UTC 21 March and 0000 UTC 22 March, and 2100 UTC 22 March). When this ETC was in its early stages around 0000 UTC 21 March, it was relatively weak with a minimum sea level pressure around 1000 hPa. While the maximum wind speed measurements near the core are around 12 m s−1 around 33°N and 140°E (Fig. 1a), LHF values were just under 400 W m−2 (Fig. 1e) while SHF was around 100 W m−2 (Fig. 1i) in this region. Along with these fluxes, we see moderate and positive air–sea temperature differences around 5°C (Fig. 1m). In addition to the largest flux values near the storm center, there is also a region of enhanced surface heat flux east of the system (around 32°N and 163°E), possibly associated with stronger winds and the advection of cooler and drier air equatorward between the low pressure center and the downstream ridge of high pressure to its southeast.
CYGNSS observations of (a)–(d) surface wind observations (m s−1), (e)–(h) LHF (W m−2), and (i)–(l) SHF (W m−2), along with (m)–(p) temperature differences between the sea surface (SST) and 10 m (T10) from MERRA-2 collocated at each specular point (°C) for an ETC in the western Pacific Ocean in March 2017. Mean sea level pressure (hPa, solid contours every 4 hPa) indicates the location of the ETC at (left) 0000 UTC 21 Mar, (left center) 2100 UTC 21 Mar, (right center) 0000 UTC 22 Mar, and (right) 2100 UTC 22 Mar. CYGNSS observations span ±1.5 h from the time listed.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
The next set of CYGNSS observations occurred between 2100 UTC 21 March and 0000 UTC 22 March. During this time, one can see that the cyclone has intensified, with a minimum sea level pressure of 980 hPa, a 20-hPa pressure decrease that, given its latitude around 32°–35°, would classify this storm as a “bomb”(Sanders and Gyakum 1980). CYGNSS observations at these times show an increase in the wind speeds to over 20 m s−1 near the core of the ETC, with winds exceeding 15 m s−1 throughout the rest of the ETC (Figs. 1b,c). Most of the areas with enhanced wind speeds also exhibit a rise in LHF and SHF, with their respective values around 400 and 200 W m−2 on the western edge of the cyclone, around 35°N and 145°E, with some parts even exceeding that (Figs. 1f,g,j,k). Additionally, we observe large LHF and SHF to the northeast of the cyclone center, around 37°N and 155°E and likely poleward of the warm front. These increases in LHF and SHF coincide with both the wind speeds and air–sea temperature differences (Figs. 1n,o) that are observed throughout most of the ETC.
While the elevated wind speeds lead to an intensification of the fluxes throughout the cyclone, we do not observe increased fluxes over most of the eastern half of the cyclone. Although LHF and SHF scale with wind speed, the air–sea differences in temperature and humidity perform an equally critical role. The southeastern portion of the ETC consists of the warm sector/warm conveyor belt, which as a result leads to lower, and even negative, air–sea temperature differences compared to the western and poleward portions of the ETC (Figs. 1n,o). These negative air–sea values, indicating that the air is warmer than the ocean surface, led to negative SHF and LHF values in this part of the ETC (around 30°N and 150°E), which indicates possible stabilization with heat and energy being transported from the lower boundary layer into the ocean surface. However, the absolute values of these fluxes (~−50 W m−2) are much lower than the positive LHF and SHF maxima observed in the western portion of the ETC. Here, colder and drier air is transported equatorward through the cold conveyor belt and possible dry intrusion behind the cold front, which then interacts with the warm and moist ocean surface beneath it (Hirata et al. 2015; Raveh-Rubin and Catto 2019). In this instance, the cooler and drier air is interacting with the warm water Kuroshio and Kuroshio Extension present in the western Pacific Ocean. These large air–sea differences and greater wind speeds of the ETC lead to the increase of LHF and SHF observed.
As this ETC evolves and moves eastward and poleward outside of CYGNSS’s range, the latest detailed observations of this cyclone at 2100 UTC 22 March continue to show wind speeds surpassing 20 m s−1 around the cyclone center, particularly on its western and equatorward sides (Fig. 1d). However, while weaker relative to the previous day, large SHF and LHF are nevertheless observed around the cold sector of the ETC (Figs. 1h,l), with a preponderance of the flux maxima correlating with the higher wind speeds. While the cyclone is farther out into the open ocean, air–sea differences were still present because of the influence of the Kuroshio Extension (Fig. 1p), as well as more poleward distribution of negative air–sea differences in the eastern half of the ETC, leading to negative LHF and SHF. While this ETC may not have been an extreme case, it does highlight CYGNSS’s ability to deliver frequent observations of low-latitude ETCs; the three overpasses presented here captured most of the surface processes associated with the cyclone and their distribution throughout the system.
b. Western Atlantic Ocean case
Nearly one year into its scientific mission, CYGNSS observed the “bomb cyclone” that occurred in January 2018 off the East Coast of the United States over the western Atlantic Ocean. This ETC experienced one of the most intense rapid periods of cyclogenesis observed in the Northern Hemisphere since 1979 (Binder et al. 2016; Hirata et al. 2019). The observations of this cyclone from the CYGNSS Fluxes were originally mentioned in a brief overview in Crespo et al. (2019); however, here we will present further aspects that were not previously discussed. CYGNSS’s first comprehensive observations of this ETC occurred around 2100 UTC 3 January 2018. At this point, it appears to be a weak-to-moderate system with minimum sea level pressure of around 1000 hPa and wind speeds under 15 m s−1 (Fig. 2a). While the wind speeds are modest throughout the system, LHF exceeding 500 W m−2 (Fig. 2d) and SHF exceeding 250 W m−2 (Fig. 2g) are already present between the core of the ETC and the coasts of South Carolina, Georgia, and Florida (around 32°N and 80°W). Air–sea temperature differences around 10°C are also observed around this area (Fig. 2j).
Similar to Fig. 1, but for an ETC in the western Atlantic Ocean in January 2018: (left) 2100 UTC 3 Jan, (center) 1500 UTC 4 Jan, and (right) 1800 UTC 4 Jan CYGNSS observations span ±1.5 h from the time listed.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
As the cyclone rapidly intensified on 4 January 2018, it quickly shifted poleward, and its core was located outside (poleward) of CYGNSS’s range during the subsequent overpasses around the 39th parallel. However, the constellation was still able to continually observe the equatorial side of the ETC on its next overpass, which featured strong winds, LHF, and SHF. Around 1500 and 1800 UTC, the minimum sea level pressure of the cyclone decreased to 964 and 960 hPa, respectively; this 36–40 hPa drop in less than 24 h met the criterion needed to be considered a rapidly intensifying bomb cyclone (Sanders and Gyakum 1980). Consistent with increased cyclone strength, CYGNSS observations show wind speeds exceeding 15 m s−1 along the outer edge of the ETC (Fig. 2b) and above 20 m s−1 closer to the cyclone center (Fig. 2c). The increased surface wind speeds lead to a significant increase in LHF and SHF, with their respective values at or exceeding 700 (Figs. 2e,f) and 300 (Figs. 2h,i) W m−2. Much like the previous case study, the local maxima of fluxes were observed in the western half of the cyclone, where colder and drier air is being transported over the warm and moist Gulf Stream, leading to large air–sea temperature differences (Figs. 2k,l). Coupled with the faster surface wind speeds, LHF and SHF increased significantly around 35°N and 75°W (Hirata et al. 2015; Raveh-Rubin and Catto 2019), leading to a significant energy transfer from the ocean into the lower atmosphere. Similar to the previous case study (Fig. 1), there is a significant distribution of negative air–sea temperature differences along the warm sector of the ETC, correlating with the negative SHF and a decrease in LHF.
While CYGNSS only had two overpasses of this rapidly developing ETC, the combined LHF and SHF observations and distribution from the first overpass correlate well with previous modeling studies of this ETC (Hirata et al. 2019, their Figs. 1a,b), with both results exhibiting combined LHF and SHF exceeding 750 W m−2. While the CYGNSS overpasses, especially the ones on 4 January, only observe the equatorward side of the ETC, these observations are valuable, as the highest fluxes associated with an extratropical cyclone are typically observed in the region trailing (west of) the cold front with high winds and large air–sea temperature and humidity differences (Neiman and Shapiro 1993).
c. Southern Hemisphere case
Analysis of the association and influence of latent and sensible heat fluxes on extratropical cyclones has often been focused on cyclones developing in the western Pacific and western Atlantic Oceans in the Northern Hemisphere. However, LHF and SHF are an important component to cyclogenesis globally. As Yu and Weller (2007) demonstrated, the highest surface heat fluxes in the Southern Hemisphere are typically observed from June to August (austral winter season), and tend to be more moderate than the Northern Hemispheric winter fluxes. One of the Southern Hemisphere ETC cases witnessed by CYGNSS occurred in August 2018 in the Southern Pacific Ocean between Australia and New Zealand with two overpasses (between 0000 and 0300 UTC 19 August and between 0000 and 0300 UTC 20 August). This ETC initially formed south of Tasmania, but later shifted equatorward, allowing for two partial overpasses from CYGNSS before it made landfall in New Zealand. Around 0000 UTC 19 August, after the ETC formed with a minimum sea level pressure of around 992 hPa, wind speeds were approximately 15 m s−1 in its northwestern sector off the southeast coast of Australia (35°S, 155°E) (Fig. 3a). Stronger winds were observed a few hours later, surpassing 20 m s−1 northwest of the cyclone center, with moderate wind speeds present throughout most of its equatorward sector (Fig. 3b). These higher wind speeds correlated with the enhanced LHF and SHF observed at this time, reaching over 400 (Figs. 3e,f) and 200 W m−2 (Figs. 3i,j), respectively. Similar to the previous case studies, the highest fluxes observed were in the western portion of the cyclone, as cooler and drier air progressed off the Australian continent and interacted with the warmer ocean surface from the East Australia Current (EAC), leading to larger air–sea temperature differences in this region (Figs. 3m,n).
As in Fig. 1, but for an ETC in the southwestern Pacific Ocean between Australia and New Zealand in August 2018: (left) 0000 UTC 19 Aug, (left center) 0300 UTC 19 Aug, (right center) 0000 UTC 20 Aug, and (right) 0300 UTC 20 Aug.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
As this ETC evolved, while the distribution of wind speeds exceeding 10 m s−1 increased on the equatorial side of the cyclone, during its next overpass CYGNSS observed lower wind speed maxima on 20 August despite a slight decrease of its minimum sea level pressure. Local wind speed maxima were observed just off the Australian coast and west of northern New Zealand (Figs. 3c,d). Despite these weaker wind speeds, higher LHF and SHF were still observed off the Australian coast along the EAC (33°S, 155°E), but with greater distribution along the Australian coast (Figs. 3g,h,k,l). This coincides with a local maxima of wind speed being located over the EAC featuring large air–sea temperature differences (Figs. 3o,p). However, given that the flux maxima are located farther from the cyclone center, it is unclear whether they had an impact on the ETC’s development. Similar to the last two case studies (Figs. 1 and 2), negative air–sea temperature differences were observed northeast of New Zealand; however, they are not as large or widespread compared to the previous cases. Although this extratropical cyclone is not as potent compared with the previous two case studies, it demonstrates that even moderate ETCs can feature significant LHF and SHF with moderate wind speeds.
These three case studies highlight CYGNSS’s capability to examine the evolution and development of some extratropical cyclones in the lower midlatitudes with recurring observations. While CYGNSS was only able to sample parts of these cyclones (with observations of the storm center postgenesis in only one case study), it was still able to observe the equatorward side of all these ETCs. Here the highest LHF and SHF typically occur due to higher wind speeds and large air–sea differences beneath the cold conveyor belt, as well as a dry intrusion behind the cold front (Hirata et al. 2015, 2016; Raveh-Rubin and Catto 2019). Given its ability to provide observations in nearly all-weather conditions with frequent revisits and multiple observations between coarse reanalysis grid points, CYGNSS can contribute greater insight into the surface processes occurring in developing ETCs. The enhanced heat fluxes correlated with the strengthening of these cyclones. The observed fluxes indicate that there are copious amounts of energy transferring from the ocean surface into the boundary layer, which could influence storm evolution. Though these observations are consistent with previous research that examined the association of strong fluxes with extratropical marine cyclogenesis, we cannot infer the precise role and impact these fluxes had solely based on these observations. Future modeling studies, similar to those found in Booth et al. (2012) and Hirata et al. (2019), will be needed to analyze the specific impact that latent and sensible heat fluxes have on these and other extratropical cyclones.
4. Composite analysis of CYGNSS ETC observations
The three case studies examined in the previous section represent just a handful of the extratropical cyclones observed by CYGNSS since its science mission began in March 2017. Approximately 36% (20%) of all ETCs that occurred within ± 40°N/S (50°N/S) were observed by CYGNSS within 500 km of their center in 2017–19. Therefore, in this section we first examine the spatial distributions of the ETCs CYGNSS could observe and then turn to compositing to explore the mean fluxes within all ETCs observed by CYGNSS.
a. Density of CYGNSS observations
Figure 4 highlights the frequency of CYGNSS specular point observations that occurred within 500 km of a low pressure center (Fig. 4a), 25 km of a warm front (Fig. 4b), and 25 km of a cold front (Fig. 4c) for all ETCs from 2017 to 2019. These constraints are similar to those used in Crespo et al. (2017), as ETC diameters are typically approximately 1000 km, with the direct surface impact from fronts frequently occurring close to the front itself (Naud et al. 2016). Despite CYGNSS’s tropical orbit, it can produce frequent observations of ETCs developing in the lower latitudes, with the frequency of observations expanding poleward in both hemispheres. In the Northern Hemisphere, the local maxima of observations occur within the western Atlantic and western Pacific basins; as mentioned in the previous section, these are regions with frequent cyclogenesis that is in part due to the influence of the Gulf Stream and Kuroshio. In the Southern Hemisphere, there is a more uniform longitudinal distribution of observations in the Pacific and Atlantic Oceans, with somewhat fewer in the southern Indian Ocean.
The number of CYGNSS specular points in 5° × 5° latitude–longitude grid cells obtained within (a) 500 km of a low pressure center, (b) 25 km of a cold front, and (c) 25 km of a warm front between 2017 and 2019. Only results over the oceans are shown.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
Although actual numbers differ, the results presented here are consistent with the prelaunch analysis performed by Crespo et al. (2017). However, while there was an abundance of observations of cold fronts and ETCs, there are significantly fewer measurements of warm fronts in all basins. With warm fronts positioned in an east–west direction directly east of the low pressure center, this means that the center of the ETC would need to be within CYGNSS’s range for a direct observation of a warm front. Recall also that the maximum distance for the warm-frontal density map is much more stringent than for the ETC density map (25 vs 500 km). CYGNSS’s orbit inclination is such that higher numbers of cold fronts are observed as they extend equatorward from the low pressure center, indicating that the ETC center does not need to be within CYGNSS’s field of view for an observation to be made. This is illustrated with the three case studies examined in the previous section, as CYGNSS observed the cold front and equatorward side of all cases but only sampled the warm front of one case study (Figs. 1 and 2).
b. Composites of wind and ocean flux observations across frontal regions
Using all cyclones in the database for which there were CYGNSS specular point measurements within 1500 km and ±3 h, we calculated the mean wind speed, LHF and SHF across cold and warm fronts. For this, we define a grid of 200-km horizontal resolution, centered on a surface front (cold or warm), and collect CYGNSS observations to populate the grid according to their distance to near coincident fronts. This is the same technique as was used in Crespo et al. (2017), which produced similar composites using prelaunch synthetic CYGNSS specular points to sample MERRA-2 winds and fluxes from the years 2014–15. The actually observed CYGNSS Flux composites (Fig. 5) resemble those prelaunch estimates (cf. Crespo et al. 2017, their Fig. 2): relatively larger LHF and SHF on the cold (west) side of the cold fronts compared to the warm (east) side, with a sharp gradient at the front (Figs. 5c,e); similarly, a relatively larger LHF/SHF on the cold side of the warm fronts compared to the warm side, and a sharp gradient across the front (Figs. 5d,f). However, there are noticeable differences in the wind composites between CYGNSS and MERRA-2; across the cold fronts, the inflexion point at the front is weaker for CYGNSS winds than MERRA-2 winds in part because of lower winds at the maximum peak on the warm side of the front (Fig. 5a). Across the warm fronts, these differences are observed in the warm sector and 200 km poleward of the warm front, where winds are the largest according to MERRA-2 (Fig. 5b). This might be related to a tendency for CYGNSS winds to be increasingly underestimated as wind speed increases above ~10 m s−1, as revealed by comparison with buoy measurements (Ruf et al. 2019a; their Fig. 4) or MERRA-2 (Balasubramaniam and Ruf 2020). That being said, an independent assessment of the CYGNSS Fluxes against buoy observations revealed that while there is a tendency for CYGNSS Fluxes to be slightly underestimated as they increase, the relative bias is small (Crespo et al. 2019). We next use the comparison between CYGNSS and MERRA-2 fluxes to examine the relative effect of changing the wind source in the flux estimate to the effect of nonuniform sampling when averaging multiple systems together.
Mean composites of (a),(b) winds, (c),(d) LHF, and (e),(f) SHF across (left) cold fronts and (right) warm fronts for ETCs within the latitude range of ±40° in 2017–19. Shown are the CYGNSS winds and fluxes (solid black line) with their reported uncertainty (blue shading), along with the MERRA-2 winds that match CYGNSS observations in time and space (dashed line).
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
While CYGNSS provides frequent sampling over the tropical and subtropical oceans, its specular point observations are nonuniform as opposed to the swaths of data from traditional remote sensing instruments. To test these sampling effects, we use MERRA-2 heat fluxes as a reference: a subset of MERRA-2 grid cells that match CYGNSS in time and space to provide a first-order evaluation of the observed winds and fluxes (hereinafter referred to as MERRA-2 @ CYGNSSS) and all MERRA-2 grid cells in the region of cold and warm fronts that are partially sampled by CYGNSS (MERRA-2 all area). A second sampling test is designed to establish the effect of CYGNSS observing just 36% of the cyclones present at any given time within the ±40° latitude range: we use MERRA-2 composites for all cyclones in this latitude range (MERRA-2 all ETC).
Regardless of which impacts are tested here, all composites differences are found to be well within CYGNSS’s estimated retrieval uncertainty (Fig. 6). Starting with LHF, across both cold and warm fronts, CYGNSS LHF is always less than MERRA-2 LHF, with a maximum in absolute difference at the location of the cold front and a minimum at the location of the warm front. Similarly, CYGNSS SHF is less than MERRA-2 SHF across both cold- and warm-frontal regions, with an absolute difference greatest slightly west of the cold fronts, as well as close to the warm front on the cold side. However, these differences are well within the estimated retrieval uncertainty obtained from the CYGNSS product files. The impact of sampling only some parts of the frontal region tends to give slightly larger LHF and lower SHF than would be obtained if CYGNSS could sample the entire frontal region (i.e., MERRA-2 @ CYGNSS minus MERRA-2 all). This impact is rather uniform across the two frontal regions and is, in absolute value, on the order of the impact of changing the wind source. Finally, the uneven sampling of the ETCs (MERRA-2 all area minus MERRA-2 all ETCs) tends to give higher fluxes where they are relatively large (cold side of both fronts) and lower there they are relatively smaller (warm side of both fronts). This suggests that the “missing” cyclones tend to be on average situated in the mean of the cyclone population. Again the absolute difference is on the order of the impact of changing the wind source and well within the retrieval uncertainty. While there are small discrepancies in terms of their magnitude, this indicates that the nonuniform sampling from CYGNSS still produces representative data of all ETCs forming in the lower latitudes (±40°).
Differences in (a),(b) LHF and (c),(d) SHF across (left) cold and (right) warm fronts for CYGNSS-observed ETCs within the latitude range of ±40° in 2017–19: CYGNSS minus MERRA-2 at same locations (MERRA2 @ CYGNSS; solid black), MERRA2 @ CYGNSS minus MERRA-2 all area (solid red) and MERRA-2 all area minus MERRA-2 all ETCs (solid blue). The dot–dashed lines mark the location of 1 standard deviation of the CYGNSS retrievals uncertainty. The vertical dotted line marks the location of the surface front, and the horizontal dotted line marks the location of the null difference.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
In the previous section, we examined three systems found in three different regions: northwest Pacific Ocean, northwest Atlantic Ocean, and in the Southern Hemisphere. Here, we explore whether these three regions show distinct characteristics in the mean from one another. Using the same database of fronts observed with CYGNSS, we divide the fronts into three basins: Northern Hemisphere Pacific Ocean (NH Pacific), Northern Hemisphere Atlantic Ocean (NH Atlantic), and all of the Southern Hemisphere (SH). We then construct the mean composites as in Fig. 8 and compare each basin to the composite obtained globally (all). Figure 7 indicates that the distribution of fluxes across cold and warm fronts is consistent in all three basins. The mean latent heat fluxes in the SH Ocean and NH Pacific Ocean are very close to the global average across both cold and warm fronts, but the NH Atlantic LHF are distinctly stronger than the other two basins and the global average (Figs. 7a,b). Note that the difference between the NH Atlantic basin and the other basins exceeds the variability (the shading in Fig. 7) expected from the varying number of fronts available between basins. This variability is estimated by randomly selecting from the entire front database (regardless of where they are) the same number of fronts as found in the NH Atlantic, composite the fluxes, and then calculate the one standard deviation across 100 such obtained composites. We calculated and plotted this one standard deviation for both the NH Atlantic and NH Pacific Oceans, which boast smaller number of fronts than the SH subset.
Mean composite of (a),(b) LHF and (c),(d) SHF across (left) cold and (right) warm fronts observed by CYGNSS, broken down by ocean basins. Solid black line: all basins. Dashed black line: Southern Hemisphere. Solid blue line: Northern Hemisphere Pacific Ocean. Solid red line: Northern Hemisphere Atlantic Ocean. The shading for the North Atlantic and North Pacific fluxes correspond to 1 standard deviation from a series of 100 composites of the same number of randomly selected transects as what is available in each basin, respectively.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
The mean sensible heat fluxes are very similar across warm fronts regardless of the basin (Fig. 7d), although the smaller Northern Hemisphere subsets tend to exhibit much wider variations on the cold side of the warm fronts. In particular, we note the wide variability caused by the limited sample size on the cold side of the warm fronts, such that in multiple places the differences across basins are not significant. In contrast, for the transects of the cold front, although SHF is very similar across basins on the warm side of the cold fronts (Fig. 7c), there are clear differences on the cold side (that exceed variability caused by sampling limitations): SHF is on average much stronger in the Northern Hemisphere than the Southern Hemisphere, and even more so for the North Atlantic Ocean than for the North Pacific Ocean within 1000 km of the front. The higher fluxes in the Northern Hemisphere than Southern Hemisphere post-cold-frontal regions are consistent with previous studies indicating that more potent LHF and SHF occur in the Northern Hemisphere (Yu and Weller 2007). Our results indicate that this is especially true for the North Atlantic. Overall, these composites generalize rather well with the three cases discussed in section 3, as the NH Atlantic case study exhibited greater surface heat fluxes compared to the SH and NH Pacific case studies. Of course, the NH Atlantic case in the previous section was an exceptional system and is not intended to be representative of all NH Atlantic cases.
c. Cyclone-centered composites of CYGNSS surface flux observations
To provide a wider context on how both LHF and SHF are distributed within the cyclone region, we also constructed cyclone-centered composites. These composites are obtained using a polar grid, by defining a grid centered on the point of minimum sea level pressure (the cyclone centers) of 100 km radial distance up to 1500 km and of 14° polar angle. In a similar fashion as what is done for the frontal transect, we accumulate CYGNSS LHF and SHF observations in each grid cell according to the location of the specular point relative to the cyclone center, and calculate the average for all cyclones with observations. Figure 8 provides the resulting composites that include all cyclones in both hemispheres for all seasons that were even partially observed by CYGNSS. Note that to ensure consistency in both hemispheres, we flip the Southern Hemisphere cyclones to have the pole side at the top of the figure in an effort to match Northern Hemisphere cyclones.
Cyclone-centered composites of (a) SHF and (b) LHF as observed with CYGNSS 2017–19 for all cyclones within ±40° latitude. Note that the Southern Hemisphere cyclones are flipped along the meridional direction to match their Northern Hemisphere counterparts such that the polar side of the cyclones is at the top of the panels. The black plus sign indicates the location of the cyclone minimum in sea level pressure.
Citation: Journal of Applied Meteorology and Climatology 60, 4; 10.1175/JAMC-D-20-0190.1
The cyclone-centered distributions of SHF (Fig. 8a) and LHF (Fig. 8b) are consistent with their cold- and warm-frontal equivalents. Both SHF and LHF are largest in the quadrant to the west of the cyclone center and are lowest in the warm sector (east and equatorward of the storm center). This matches the relatively strong contrast across cold fronts. We also note that LHF minimum follows the region of the cyclone where the warm fronts are found on average (due east from the storm center). It is noteworthy that the composite accumulation is consistent with the picture we obtained using observations of individual cyclones (section 3). To our knowledge, this is the first time the climatological mean of surface fluxes within extratropical cyclones has been obtained in this way using spaceborne observations of cyclones within ±40° latitude.
5. Discussion and conclusions
While CYGNSS is a tropical mission with a focus on tropical cyclones, this paper illustrated its ability to observe the surface processes within and around marine extratropical cyclones found in the lower latitudes (±40°). CYGNSS provides valuable information where data from in situ measurements and other remote sensing data may be lacking with its all-weather observations and rapid revisit, observing over a third of all ETCs occurring over the tropical and subtropical oceans. As exhibited in section 3, CYGNSS was capable of observing surface winds and heat fluxes within and around these ETCs as they matured and strengthened. While it was not able to obtain observations poleward of the ETC center in two of the three case studies, CYGNSS was still able to provide significant details of surface winds, LHF, and SHF on the equatorward side of these cyclones. The strongest combined surface heat fluxes are frequently observed in this portion of the ETC as the cold conveyor belt and dry intrusion behind the cold front commonly interact with the warm and moist ocean surface (Raveh-Rubin and Catto 2019). The faster wind speeds and large air–sea differences lead to more potent LHF and SHF. Robust combined surface heat fluxes were observed in the rapidly developing systems in the western Atlantic and western Pacific Oceans (Figs. 1 and 2); however, even modest values of LHF and SHF were present in the weaker ETC observed in the South Pacific (Fig. 3).
The presented case studies indicate the role surface processes may perform in extratropical cyclones. These results are compatible with preceding observational and modeling studies that revealed the role surface fluxes could play on ETC development (Booth et al. 2012; Small et al. 2014; Hirata et al. 2015, 2018, 2019). The warm conveyor belt transports a majority of the moisture that is transformed into clouds and precipitation within an ETC (Harrold 1973; Browning et al. 1973; Carlson 1980; Browning 1986; Wernli 1997; Catto 2016). Given that WCBs consistently originate near the surface within the marine boundary layer (Wernli and Davies 1997; Eckhardt et al. 2004; Madonna et al. 2014), they can transport considerable quantities of LHF and SHF into an ETC (Browning 1990). Considering most WCBs begin their ascents between approximately 20°–40° in latitude in both hemispheres, CYGNSS can contribute invaluable surface observations at their source regions.
The observational analysis presented here provides information about some of the connections that may exist between ocean surface heat fluxes and the dynamics and structures of extratropical cyclones. However, this observational analysis alone merely shows correlation of fluxes and ETC development. It does not imply causation of changes in ETC genesis and evolution due to the presence of copious amounts of combined surface heat fluxes. Explorations of the mechanisms related to the influence of LHF and SHF on ETC development can be accomplished using a model as a numerical laboratory. The observational analysis presented here will provide a foundation for developing a modeling analysis that utilizes CYGNSS’s wind and flux observations to better understand the direct impact LHF and SHF may have on ETC synoptic and mesoscale structure, dynamics, clouds, and precipitation.
Although this paper concentrated on just three case studies, CYGNSS can provide observations of about one-third of all extratropical cyclones that develop in the lower midlatitudes. The composite analysis indicates that a majority of CYGNSS’s ETC observations occur in the subtropical western Atlantic and western Pacific Ocean basins in the Northern Hemisphere, as well as throughout various parts of the Southern Hemisphere (Fig. 4). These results were consistent with prelaunch expectations (Crespo et al. 2017), although the number of warm-frontal observations was lower than expected, possibly related to interannual variability as well as being located poleward of most CYGNSS observations. The impact of CYGNSS’s nonuniform sampling, as compared with traditional remote sensing instruments, was found be relatively small, especially relative to retrieval uncertainty, and to be of the same order as the variability caused by changing the wind source when calculating the fluxes (Figs. 5 and 6). A composite analysis across fronts and cyclone-centered demonstrated that CYGNSS can provide detailed composites of LHF and SHF, including when broken down by basins (NH Atlantic/Pacific and Southern Hemisphere Oceans) (Fig. 7). These results correlated with the case studies presented and previous modeling studies, with strong contrast in flux magnitudes across the equator side of the cyclones, that is, across the cold fronts. On average, CYGNSS Fluxes composites indicated comparatively strong combined surface heat fluxes in the Atlantic Basin in the Northern Hemisphere.
CYGNSS’s observations of extratropical cyclones complement existing in situ data and conventional remote sensing platforms; however, CYGNSS has its shortcomings. While it can observe a notable abundance of ETCs, CYGNSS observations are confined to low latitudes because of its tropical orbit. Additionally, there are ambiguities present in CYGNSS’s wind and surface heat flux products (Crespo et al. 2019) at larger values, which may limit the utility of its observations of the most intense ETCs, though these uncertainties are presumed to improve with subsequent data product releases. Additionally, the CYGNSS Surface Heat Flux Product relies on reanalysis data for the thermodynamic variables needed to estimate LHF and SHF. Notwithstanding these caveats, CYGNSS can serve as a comprehensive tool in better understanding the surface processes in and throughout extratropical cyclones, as CYGNSS’s detailed wind speed observations in all-weather conditions can provide greater insight to surface processes occurring at smaller scales and in rapidly changing conditions. Likewise, ETCs that form and are predominantly located outside of CYGNSS’s orbit and field of view, often derive much of their energy and moisture from tropical and subtropical regions. While this paper investigated ETCs directly observed by CYGNSS, its numerous measurements of the tropical and subtropical oceans could advance our comprehension of marine-based extratropical cyclones, their source regions, and other high-impact weather phenomena, such as atmospheric rivers or tropical moisture exports (e.g., Ralph et al. 2018). Forthcoming enhancements to CYGNSS and potential deployments of additional remote sensing platforms will continue to provide enhanced observations of surface processes and smaller-scale features that ensue within and throughout extratropical cyclones.
Acknowledgments
This work was supported by NASA CYGNSS Science Team Grant NNH17ZDA001N and by the CYGNSS mission under NASA Science Mission Directorate Contract NNL13AQ00C. The detailed comments and suggestions given by three anonymous reviewers were helpful and greatly appreciated. This work was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. Government sponsorship is acknowledged.
Data availability statement
A description of and access to the extratropical cyclone database described and used in sections 2 and 4 is available online (https://data.giss.nasa.gov/storms/obs-etc/). The database is arranged as follows: for each cyclone in the database, a file is created that includes storm center information (time, longitude, latitude, and SLP at the center) as well as warm- and cold-frontal locations estimated from three different methods each. The storm files are arranged into yearly tar files and are available online (https://portal.nccs.nasa.gov/datashare/Obs-ETC/Fronts-ETC). The CYGNSS Level-2 Wind Speed (CDR V1.0) and Surface Heat Flux (CDR V1.0) data were obtained online through NASA’s Physical Oceanography Distributed Active Archive Center (PO.DAAC) (https://podaac.jpl.nasa.gov/CYGNSS?sections=data).
REFERENCES
Balasubramaniam, R., and C. Ruf, 2020: Neural network based quality control of CYGNSS wind retrieval. Remote Sens., 12, 2859, https://doi.org/10.3390/rs12172859.
Bauer, M., and A. D. Del Genio, 2006: Composite analysis of winter cyclones in a GCM: Influence on climatological humidity. J. Climate, 19, 1652–1672, https://doi.org/10.1175/JCLI3690.1.
Bauer, M., G. Tselioudis, and W. B. Rossow, 2016: A new climatology for investigating storm influences in and on the extratropics. J. Appl. Meteor. Climatol., 55, 1287–1303, https://doi.org/10.1175/JAMC-D-15-0245.1.
Bentamy, A., K. B. Katsaros, A. M. Mestas-Nunez, W. M. Drennan, E. B. Forde, and H. Roquet, 2003: Satellite estimates of wind speed and latent heat flux over the global oceans. J. Climate, 16, 637–656, https://doi.org/10.1175/1520-0442(2003)016<0637:SEOWSA>2.0.CO;2.
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., 1919: On the structure of moving cyclones. Mon. Wea. Rev., 47, 95–99, https://doi.org/10.1175/1520-0493(1919)47<95:OTSOMC>2.0.CO;2.
Bjerknes, J., and H. Solberg, 1922: Life cycle of cyclones and the polar front theory of atmospheric circulation. Geofys. Publ., 3, 1–18.
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.
Branscome, L. E., W. J. Gutowski, and D. A. Stewart, 1989: Effects of surface fluxes on the nonlinear development of baroclinic waves. J. Atmos. Sci., 46, 460–475, https://doi.org/10.1175/1520-0469(1989)046<0460:EOSFOT>2.0.CO;2.
Browning, K. A., 1986: Conceptual models of precipitation systems. Wea. Forecasting, 1, 23–41, https://doi.org/10.1175/1520-0434(1986)001<0023:CMOPS>2.0.CO;2.
Browning, K. A., 1990: Organization of clouds and precipitation in extratropical cyclones. Extratropical Cyclones: The Erik Palmén Memorial Volume, C.W. Newton and E.O. Holopainen, Eds., Amer. Meteor. Soc., 129–153.
Browning, K. A., M. E. Hardman, T. W. Harrold, and C. W. Pardoe, 1973: Structure of rainbands within a mid-latitude depression. Quart. J. Roy. Meteor. Soc., 99, 215–231, https://doi.org/10.1002/qj.49709942002.
Carlson, T. N., 1980: Airflow through midlatitude cyclones and the comma cloud pattern. Mon. Wea. Rev., 108, 1498–1509, https://doi.org/10.1175/1520-0493(1980)108<1498:ATMCAT>2.0.CO;2.
Catto, J. L., 2016: Extratropical cyclone classification and its use in climate studies. Rev. Geophys., 54, 486–520, https://doi.org/10.1002/2016RG000519.
Catto, J. L., C. Jakob, G. Berry, and N. Nicholls, 2012: Relating global precipitation to atmospheric fronts. Geophys. Res. Lett., 39, L10805, https://doi.org/10.1029/2012GL051736.
Cione, J. J., S. Raman, and L. J. Pietrafesa, 1993: The effect of Gulf Stream-induced baroclinicity on U.S. East Coast winter cyclones. Mon. Wea. Rev., 121, 421–430, https://doi.org/10.1175/1520-0493(1993)121<0421:TEOGSI>2.0.CO;2.
Clarizia, M. P., C. Gommenginger, S. Gleason, M. Srokosz, C. Galdi, and M. di Bisceglie, 2009: Analysis of GNSS-R delay-Doppler maps from the UK-DMC satellite over the ocean. Geophys. Res. Lett., 36, L02608, https://doi.org/10.1029/2008GL036292.
Clarizia, M. P., C. S. Ruf, P. Jales, and C. Gommenginger, 2014: Spaceborne GNSS-R minimum variance wind speed estimator. IEEE Trans. Geosci. Remote Sens., 52, 6829–6843, https://doi.org/10.1109/TGRS.2014.2303831.
Clarizia, M. P., V. Zavorotny, and C. Ruf, 2018: Level 2 wind speed retrieval. Algorithm Theoretical Basis Doc., CYGNSS Project Doc. 148-0138, Rev. 5, 96 pp., https://clasp-research.engin.umich.edu/missions/cygnss/reference/148-0138%20ATBD%20L2%20Wind%20Speed%20Retrieval_Rev5_Aug2018_release.pdf.
Crespo, J. A., D. J. Posselt, C. M. Naud, and C. D. usy-Virat, 2017: Assessing CYGNSS’s potential to observe extratropical fronts and cyclones. J. Appl. Meteor. Climatol., 56, 2027–2034, https://doi.org/10.1175/JAMC-D-17-0050.1.
Crespo, J. A., D. J. Posselt, and S. Asharaf, 2019: CYGNSS surface heat flux product development. Remote Sens., 11, 2294, https://doi.org/10.3390/rs11192294.
CYGNSS, 2020a: CYGNSS Level 2 Ocean Surface Flux Climate Data Record. Version 1.0. PO.DAAC, accessed 1 November 2020, https://doi.org/10.5067/CYGNS-C2H10.
CYGNSS, 2020b: CYGNSS Level 2 Climate Data Record. Version 1.0. PO.DAAC, accessed 1 July 2020, https://doi.org/10.5067/CYGNS-L2C10.
de Vries, H., S. Scher, R. Haarsma, S. Drijfhout, and A. van Delden, 2019: How Gulf-Stream SST-fronts influence Atlantic winter storms: Results from a downscaling experiment with HARMONIE to the role of modified latent heat fluxes and low-level baroclinicity. Climate Dyn., 52, 5899–5909, https://doi.org/10.1007/s00382-018-4486-7.
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.
Edson, J., and Coauthors, 2013: On the exchange of momentum over the open ocean. J. Phys. Oceanogr., 43, 1589–1610, https://doi.org/10.1175/JPO-D-12-0173.1.
Gelaro, R., and Coauthors, 2017: The Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2). J. Climate, 30, 5419–5454, https://doi.org/10.1175/JCLI-D-16-0758.1.
Harrold, T. W., 1973: Mechanisms influencing the distribution of precipitation within baroclinic disturbances. Quart. J. Roy. Meteor. Soc., 99, 232–251, https://doi.org/10.1002/qj.49709942003.
Hawcroft, M., L. Shaffrey, K. Hodges, and H. Dacre, 2012: How much Northern Hemisphere precipitation is associated with extratropical cyclones? Geophys. Res. Lett., 39, L24809, https://doi.org/10.1029/2012GL053866.
Heideman, K. F., and J. M. Fritsch, 1988: Forcing mechanisms and other characteristics of significant summertime precipitation. Wea. Forecasting, 3, 115–130, https://doi.org/10.1175/1520-0434(1988)003<0115:FMAOCO>2.0.CO;2.
Hewson, T. D., 1998: Objective fronts. Meteor. Appl., 5, 37–65, https://doi.org/10.1017/S1350482798000553.
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.
Hirata, H., R. Kawamura, M. Kato, and T. Shinoda, 2016: Response of rapidly developing extratropical cyclones to sea surface temperature variations over the western Kuroshio–Oyashio confluence region. J. Geophys. Res. Atmos., 121, 3843–3858, https://doi.org/10.1002/2015JD024391.
Hirata, H., R. Kawamura, M. Kato, and T. Shinoda, 2018: A positive feedback process related to the rapid development of an extratropical cyclone over the Kuroshio/Kuroshio Extension. Mon. Wea. Rev., 146, 417–433, https://doi.org/10.1175/MWR-D-17-0063.1.
Hirata, H., R. Kawamura, M. Nonaka, and K. Tsuboki, 2019: Significant impact of heat supply from the Gulf Stream on a “super bomb” cyclone in January 2018. Geophys. Res. Lett., 46, 7718–7725, https://doi.org/10.1029/2019GL082995.
Hotta, D., and H. Nakamura, 2011: On the significance of the sensible heat supply from the ocean in the maintenance of the mean baroclinicity along storm tracks. J. Climate, 24, 3377–3401, https://doi.org/10.1175/2010JCLI3910.1.
Liu, W. T., K. B. Katsaros, and J. A. Businger, 1979: Bulk parameterization of the air–sea exchange of heat and water vapor including the molecular constraints at the interface. J. Atmos. Sci., 36, 1722–1735, https://doi.org/10.1175/1520-0469(1979)036<1722:BPOASE>2.0.CO;2.
Madonna, E., H. Wernli, H. Joos, and O. Martius, 2014: Warm conveyor belts in the ERA-Interim dataset (1979–2010). Part I: Climatology and potential vorticity evolution. J. Climate, 27, 3–26, https://doi.org/10.1175/JCLI-D-12-00720.1.
Mak, M., 1998: Influence of surface sensible heat flux on incipient marine cyclogenesis. J. Atmos. Sci., 55, 820–834, https://doi.org/10.1175/1520-0469(1998)055<0820:IOSSHF>2.0.CO;2.
Mears, C., D. Smith, and F. Wentz, 2000: Detecting rain with QuikSCAT. Proc. Int. Geoscience and Remote Sensing Symp., Honolulu, HI, IEEE, 1235–1237, https://doi.org/10.1109/IGARSS.2000.858078.
Naud, C. M., A. D. Del Genio, M. Bauer, and W. Kovari, 2010: Cloud vertical distribution across warm and cold fronts in CloudSat–CALIPSO data and a general circulation model. J. Climate, 23, 3397–3415, https://doi.org/10.1175/2010JCLI3282.1.
Naud, C. M., J. F. Booth, and A. D. Del Genio, 2016: The relationship between boundary layer stability and cloud cover in the post-cold-frontal region. J. Climate, 29, 8129–8149, https://doi.org/10.1175/JCLI-D-15-0700.1.
Neiman, P. J., and M. A. Shapiro, 1993: The life cycle of an extratropical marine cyclone. Part I: Frontal-cyclone evolution and thermodynamic air–sea interaction. Mon. Wea. Rev., 121, 2153–2176, https://doi.org/10.1175/1520-0493(1993)121<2153:TLCOAE>2.0.CO;2.
Pfahl, S., and H. Wernli, 2012: Quantifying the relevance of cyclones for precipitation extremes. J. Climate, 25, 6770–6780, https://doi.org/10.1175/JCLI-D-11-00705.1.
Portabella, M., A. Stoffelen, W. Lin, A. Turiel, A. Verhoef, J. Verspeek, and J. Ballabera-Poy, 2012: Rain effects on ASCAT-retrieved winds: Toward an improved quality control. IEEE Trans. Geosci. Remote Sens., 50, 2495–2506, https://doi.org/10.1109/TGRS.2012.2185933.
Ralph, F. M., M. D. Dettinger, M. M. Cairns, T. J. Galarneau, and J. Eylander, 2018: Defining “atmospheric river”: How the Glossary of Meteorology helped resolve a debate. Bull. Amer. Meteor. Soc., 99, 837–839, https://doi.org/10.1175/BAMS-D-17-0157.1.
Raveh-Rubin, S., and J. 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.
Richter, D. H., and D. P. Stern, 2014: Evidence of spray-mediated air-sea enthalpy flux within tropical cyclones. Geophys. Res. Lett., 41, 2997–3003, https://doi.org/10.1002/2014GL059746.
Ruf, C. S., and Coauthors, 2016: New ocean winds satellite mission to probe hurricanes and tropical convection. Bull. Amer. Meteor. Soc., 97, 385–395, https://doi.org/10.1175/BAMS-D-14-00218.1.
Ruf, C. S., S. Asharaf, R. Balasubramaniam, S. Gleason, T. Lang, D. McKague, D. Twigg, and D. Waliser, 2019a: In-orbit performance of the constellation of CYGNSS hurricane satellites. Bull. Amer. Meteor. Soc., 100, 2009–2023, https://doi.org/10.1175/BAMS-D-18-0337.1.
Ruf, C., S. Gleason, and D. S. McKague, 2019b: Assessment of CYGNSS wind speed retrieval uncertainty. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 12, 87–97, https://doi.org/10.1109/JSTARS.2018.2825948.
Sanders, F., and J. R. Gyakum, 1980: Synoptic-dynamic climatology of the “bomb.” Mon. Wea. Rev., 108, 1589–1606, https://doi.org/10.1175/1520-0493(1980)108<1589:SDCOT>2.0.CO;2.
Schemm, S. I. Rudeva, and I. Simmonds, 2015: Extratropical fronts in the lower troposphere—Global perspectives obtained from two automated methods. Quart. J. Roy. Meteor. Soc., 141, 1686–1698, https://doi.org/10.1002/qj.2471.
Shapiro, M. A., and D. Keyser, 1990: Fronts, jet streams, and the tropopause. Extratropical Cyclones: The Erik Palmén Memorial Volume, C. W. Newton and E. Holopainen, Eds., Amer. Meteor. Soc., 167–191.
Simmonds, I., K. Keay, and J. A. T. Bye, 2012: Identification and climatology of Southern Hemisphere mobile fronts in a modern reanalysis. J. Climate, 25, 1945–1962, https://doi.org/10.1175/JCLI-D-11-00100.1.
Small, R. J., R. A. Tomas, and F. O. Bryan, 2014: Storm track response to ocean fronts in a global high-resolution climate model. Climate Dyn., 43, 805–828, https://doi.org/10.1007/s00382-013-1980-9.
Vukovich, F. M., J. W. Dunn, and B. W. Crissman, 1991: Aspects of the evolution of the marine boundary layer during cold-air outbreaks off the southeast coast of the United States. Mon. Wea. Rev., 119, 2252–2279, https://doi.org/10.1175/1520-0493(1991)119<2252:AOTEOT>2.0.CO;2.
Wernli, H., 1997: A Lagrangian-based analysis of extratropical cyclones. Part 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.
Yu, L. S., and R. A. Weller, 2007: Objectively analyzed air–sea heat fluxes for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88, 527–540, https://doi.org/10.1175/BAMS-88-4-527.