FU Dan, LI Pengyuan, and FU Gang

Department of Meteorology, Ocean University of China, Qingdao 266100, P. R. China

An Observational and Modeling Study of Extratropical Transition of Hurricane Sandy in 2012

FU Dan, LI Pengyuan, and FU Gang*

Department of Meteorology, Ocean University of China, Qingdao 266100, P. R. China

? Ocean University of China, Science Press and Springer-Verlag Berlin Heidelberg 2015

Around 30 October 2012, Hurricane Sandy made landfall along the New Jersey shoreline after its completion of extratropical transition and transformation into an extratropical cyclone. The strong gale induced a catastrophic storm surge, and caused 72 death and damage of more than $50 billion. In this paper, the evolutionary process and spatial structure of the Hurricane Sandy during its extratropical transition were investigated by using Weather Research and Forecasting (WRF) version 3.3.1 modeling results and National Center for Environmental Prediction (NCEP) Coupled Forecast System model version 2 reanalysis datasets (CFSv2). It is found that during the upper-level trough interaction on 29 October, Sandy gradually fused with a pre-existing mid-latitude low-pressure system, and finished the re-intensification. WRF modeling results showed that the second peak occurred mainly due to the enhanced vertical motion, reduced vertical wind shear as well as the supplement of potential vorticity resulting from trough interaction over the southeast of Great Lakes. The cold continental air from the back of trough was encircled within the warm core system cyclonically, forming the characteristic of warm seclusion.

Hurricane Sandy; extratropical transition; mid latitude trough; WRF modeling

1 Introduction

In meteorological community, Hurricane Sandy from 22 October to 31 October 2012 has widely been regarded to be a classic late-season tropical cyclone (TC) over the North Atlantic, as well as the deadliest and most destructive hurricane in the 2012 Atlantic hurricane season. During its movement crossing seven countries, at least 286 people were killed. On 29 October 2012, a devastating storm surge induced by the Hurricane Sandy attacked the United States stretching from the southern New Jersey to Rhode Island, causing 72 deaths and approximately $50 billion damage (Blake et al., 2013).

Previous studies indicated that some of TCs might be transformed into extratropical cyclones at the end of their lives, usually occurring in mid-latitudes, where there is sufficient forcing from the upper-level troughs (Wang et al., 2012; DiMego and Bosart, 1982; Hart and Evans, 2001; Klein et al., 2000). This process in which TCs lose their tropical characteristics and become extratropical cyclones is usually defined as extratropical transition (ET) (Sekioka, 1970, 1972; Klein et al., 2000; Ritchie and Elsberry, 2001; Jones et al., 2003). During ET, the cyclone gradually loses its warm core structure and is distorted in shape, becoming less symmetric in circulation,convection, and humidity, which results in strong winds, extremely heavy rainfall as well as other potential secondary disasters like flooding and mudslide (Evans and Hart, 2008). Although in hurricane seasonextratropical transition of TC has been documented over the Northwest Pacific, Northeast Pacific, North Atlantic, and Southeast Indian Ocean, the systematic investigation of ET in later hurricane season is rare.

Hurricane Sandy underwent ET process and became an extratropical cyclone around 21 UTC 29 October 2012 (Blake et al., 2013). During its ET process, Sandy produced torrential rainfalls that mainly fell into the south and west of Sandy’s moving track, which is consistent with previous studies (Bergeron, 1954; Palmén, 1958; Carr and Bosart, 1978; Foley and Hanstrum, 1994). The extreme precipitation was reported in Bellevue, Maryland with a peak amount of 325.88 mm, and an average of 127–178 mm of rainfall in the east of Maryland and Virginia, the south of Delaware and New Jersey. Due to the deep convection, even a snowfall occurred in an area stretching from North Carolina to the southwestern Pennsylvania, and the amount being over 90 cm in local accumulation. Additionally, the ET process of Sandy also was accompanied with a catastrophic storm surge along the northeastern United States coastline driven by the sharp pressure gradient and strong surface easterly flow.

Several previous studies have been made to investigate the ET process of hurricane. Evans and Hart (2003) indicated that although cold core is the predominant factor tojudge the completion of ET for some cases, sometimes, the cold core is not so obvious even in the extratropical cyclone situation. Thorncroft and Jones (2000) found that a cold core develops at low levels while a weak warm core is always maintained above 600 hPa during the ET of Hurricane Felix in 1995. Browning et al. (1998) investigated the ET of Hurricane Lili in 1996, and illustrated how Lili developed into a system with warm core seclusion structure, which was similar to the final phase of extratropical frontal cyclone studied by Shapiro and Keyser (1990). Browning (2004) also indicated that the TCs with warm seclusion phase might easily produce damage near sea surface.

For Hurricane Sandy, it reached a second peak during its ET process associated with warm seclusion stage, and the upper-level warm core was still maintained even when being a mature extratropical cyclone. The present paper aims to use the observational data as well as the simulation results from the Weather Research and Forecasting (WRF) Model to make a systematic study about the evolution process of Hurricane Sandy’s extratropical transition, including how Sandy loses its tropical structure and fuses with a pre-existing mid-latitude upper-level trough, as well as how Sandy is re-intensified during the warm seclusion phase. The paper is organized as follows. Section 2 displays an overview of Hurricane Sandy’s life cycle. Section 3 introduces the data and methodology. Section 4 observationally and numerically depicts the interaction between Sandy’s warm core and the midlatitude low-pressure system. Finally, Section 5 presents the conclusions.

2 Overview of Hurricane Sandy

Hurricane Sandy initially formed over the southern part of Kingston, Jamaica (13.3?N, 78.2?W) from an easterly wave which moved westward on 22 October, 2012. At that time, it was first identified as a tropical depression (TD) by National Hurricane Center (NHC) (Blake et al., 2013). After then, Sandy moved westward and developed into a tropical storm around 1800 UTC 22 October, being contributed by the slight vertical wind shear and warm SST. This system strengthened into a category-3 (Simpson) hurricane with about 954 hPa minimum central sea level pressure and 50 m s-1surface wind speed slightly before it went northward and made landfall in San Diego, Cuba around 0525 UTC 25 October. Meanwhile, a well- defined‘eye’ structure with deep eyewallcould be observed from the satellite image (Figs.1a and 1c). Sandy moved northwestward and was downgraded to a tropical storm soonafter crossing Cuba around 0000 UTC 27 October, due to the upper-level cold and dry air mass accompanied with strong southwesterly wind shear (Michael, 2012), with the convection becoming asymmetric (Fig.1a).

Fig.1 (a), (b) Hurricane Sandy track from HURDAT2 overlaid by CIRA-RAMMB 4-km remapped enhanced infrared satellite imagery (shaded; unit: ℃) at 0625 UTC 25, 2345 UTC 25, 1825 UTC 28 and 0015 UTC 30 October respectively. The Sandy centers are marked by symbols of ‘L’,‘L’ for tropical depression, tropical storm, hurricane an d e xtratropical low pressure system according to HURDAT2. (c) Time series of HURDAT2 minimum central seapressure (SLP;solid line with dots; unit: hPa), maximum sustained wind speed (solid line with circles; unit:This figure isadapted from the Fig.1 of Galarneau et al. (2013).

As Sandy turned northeastward, a well defined vortex structure was reconstructed and the convection center was mainly located on Sandy’s northwest flank by 0000 UTC 28 October (Fig.1b). During the day-time of October 29, Sandy approached a second pre-existing upper-level trough, and moved to northwestward towards the northeastern United States shoreline due to the midtropospheric blocking high pressure pattern. In the meantime, the diameter of whole Sandy system reached 1850 km, and the storm was re-intensified into a category-2 hurricane with 940 hPa minimum central sea level pressure and 45 m s-1wind speed at 1800 UTC 29 October (Fig.1c). The rapid intensification of wind speed and sea level pressure gradient drove a devastating storm surge that caused damage of tens of billion dollars. On 5 November 2012, NHC ranked the Hurricane Sandy as the second costliest US hurricane since 1900, just behind Hurricane Katrina in 2005.

3 Data and Modeling Results

3.1 Observational and Reanalysis Data

In this paper, we employ the following data to study the ET process of Hurricane Sandy systematically.

1) Hurricane data 2 (HURDAT2) offered by NHC, which provides the hurricane’s best track position, minimum sea level pressure, and maximum sustained wind speed. This data is available at 0000, 0600, 1200, 1800 UTC every day. Available from: http://www.nhc.noaa.gov/ data/ hurdat/hurdat2-1851-2013-052714.txt.

2) 4 km×4 km resolution Remapped Color Enhanced Infrared Imagery from the Cooperative Institute for Research in the Atmosphere (CIRA, Colorado State University)– Regional and Mesoscale Meteorology Branch (RAMMB). Available from: http://rammb.cira.colostate.edu/products/ tc_realtime/.

3) Multiplatform Satellite Surface Wind Analysis product, which combines the information from different data sources including Advanced Microwave Sounding Unit (AMSU) winds, Cloud-drift/IR/WV winds, IR-proxy winds and Scatterometer winds. Available from: http://rammb.cira. colostate.edu/products/tc_realtime/.

4) Sounding data of station 72501 downloaded from University of Wyoming, which is available 2 times per day. Available from: http://weather.uwyo.edu/upperair/sounding. html.

5) Climate Forecast System version 2 (CFSv2) data from NCEP (National Center for Environmental Prediction), the spatial resolution is 0.5?×0.5? and 37 layers in the vertical direction (1000, 975, 950, 925, 900, 875, 850, 825, 800, 775, 750, 700, 650, 600, 550, 500, 450, 400, 350, 300, 250, 225, 200, 175, 150, 125, 100, 70, 50, 30, 20, 10, 7, 5, 3, 2 , 1 hPa), the time interval is 6 hours. Available from: http://rda.ucar.edu/datasets/ds094.0/.

6) Real-time, global, sea surface temperature (RTG_ SST_HR) data from NCEP, the spatial resolution is 0.083?× 0.083?, and it is a daily mean product. Available from: ftp://polar.ncep.noaa.gov/pub/history/sst.

3.2 Modeling Results

The numerical simulation of ET process of Hurricane Sandy was performed by WRFv3.3.1 model. Two domains were used in the simulation with an outer domain of 21 km×21 km horizontal resolution which covers the whole east of United States and west of Atlantic basin, and an inner nest domain of 7 km×7 km (Fig.2a). The total integration times for outer domain and inner domain were 60 hours (initialized at 0000 UTC 28 October 2012) and 48 hours (initialized at 1200 UTC 28 October 2012), respectively. Both two domains had 40 layers vertically, up to 50 hPa. For the configuration, Yonsei University Planenary Boundary Layer (PBL) scheme (YSU; Noh et al., 2003; Hong et al., 2006), Noah-MP Land surface physics (Chen and Dudhia, 2001), Dudhia (1989) short wave radiation scheme, Rapid Radiative Transfer Model (RRTM) long wave radiation scheme (Mlawer et al., 1997), Betts-Miller-Janjic cumulus parameterization (Betts, 1986; Betts and Miller, 1986; Janjic, 1994) and WDM 6-class microphysics scheme (Lin et al., 1983) were used in the simulation (Table 1).

The simulation results of inner domain during the ET period of Hurricane Sandy are shown in Fig.2. Generally, the track simulation is relative accurate, with less than 30 km deviation compared with the position from HURDAT2 data. Basically, the nested simulation of WRFv3.3.1 is able to capture the interaction between Sandy and second upper-level trough as well as Sandy’s northwestward steering toward New Jersey shoreline. However, the intensity simulation is not as good as track simulation. This is perhaps due to the impotency of initial condition, neither simulated wind speed nor central sea level pressure was robust enough at first 3-h running. After a period of adjustment, the WRF model successfully captured Sandy’s second peak and rapid dissipation both in pressure and wind speed field. But the simulated wind speed at 10-m height was roughly 2–7 m s-1smaller than the observations (Fig.2b). The comparison of the 10-m wind speed between the satellite data and WRF simulation at 1800 UTC 29 October is shown in Figs.2c and 2d. WRF model could reconstruct the Sandy’s circulation as well as the‘eye’ structure during the its secondary peak, but the area with more than 50 knot wind speed seemed to be too large both at the south and the north of Sandy’s center, while the wind speed at the south of 35?N was too weak.

As the initial condition was merely derived from CFSv2 reanalysis data, which may have some biases with observation, the simulated intensity is less accurate compared with the track simulation. Anyhow, the WRFv3.3.1 model is able to reproduce the whole ET process of Sandy successfully. It is hoped in future that data assimilation and ensemble simulation will improve the simulation accuracy.

Fig.2 (a) Comparison of Hurricane Sandy track between WRF simulation (red solid line with dot) and HURDAT2 (black solid line with dot). The nested domain used for the simulation is also shown (red dash box). (b) Time series of HURDAT2 minimum central sea level pressure (SLP; black solid line with dots; unit: hPa), intensity (Vmax; black solid line with circles; unit: m s-1) as well as WRF simulated minimum central sea level pressure (red line; unit: hPa) and 10 m wind speed (green line; unit: m s-1) of Hurricane Sandy. (c) Multiplatform Satellite Surface Wind Analysis at 1800 UTC 29 October (unit: knot). (d) WRF simulated 10 m wind speed (unit: knot) and barb at 1800 UTC 29 October.

Table 1 Model parameterization for Hurricane Sandy (2012) simulation

4 Analyses of Sandy’s ET Process

4.1 Interaction with Mid-Latitude Trough

When Sandy moved poleward to mid-latitude, the environmental conditions changed a lot, characterized by the enhancement of Coriolis force, decreasing of sea surface temperature as well as increased atmospheric baroclinicity. A pre-existing upper-level trough deepened due to the building of mid-tropospheric trough over the central Atlantic basin and located over the central America from 28 to 30 October. Additionally, the mid-latitudetrough dove southeastward from central America and formed a negative tilt late on 28 October. Under this condition, northwestward moving Sandy gradually interacted with this upper-tropospheric trough. As Sandy was located downstream, the upper-level trough not only provided great divergence but also decreased the vertical wind shear, especially at the west flank of Hurricane Sandy (Figs.3a and 4a). Increased upper-level divergence in conjunction with decreased vertical wind shear further enhanced vertical motion near the Sandy’s center, making Sandy more convective. At this time, the Sandy’s central air mass mean temperature was distinctly higher than the surrounding.

Around 0600 UTC 29 October, Sandy moved closer to the upper-level trough. At the 500 hPa level, Sandy’s 5600 gpm isobar had been fused with mid-latitude trough, and deep convection near Sandy’s warm core center and northwest flank wasmore concentrated (Figs.3b and 4b). Late on 29 October, a new but smaller low-pressure system was drawn out from negative tilt trough at 200 hPa and approached Sandy (Fig.3c). Simultaneously, the cold continental air from the back of mid-latitude tough intruded southeastward and entered Sandy’s warm core cyclonically prior to its landfall along New Jersey shoreline (Figs.3c and 3d). At mid-level, Sandy’s isobars had fused with mid-latitude low-pressure system, forming a new and big low-pressure center and even being deepened.

Using sea surface and satellite observations, we may review this period more clearly. At 1500 UTC 29 October, an expanded low-pressure system was observed at surface, with a stationary front and an occlusion located on the western and northern sides of Sandy, respectively (Fig.5a). A dense overcast near cyclone center could still be found from the satellite image at this time, but disappeared in the next 6 hours (Fig.5b). By 2100 UTC 29 October, the latter cold front caught up with the former warm front largely, and occlusion was even wrapped into the cyclone center. Moreover, Sandy’s warm core was lifted by the latter cold air and further secluded, which made Sandy acquire the characteristics of a warm seclusion type cyclone (Galarneau et al., 2013). As Sandy no longer maintained its deep convection and organized frontal structure, it could be confirmed that Sandy had become an extratropical cyclone after finishing its ET process by the end of 29 October.

Twelve hours later, the upward vertical motion near the Sandy’s center became very weak, and Sandy had totally fused with pre-existing upper-level trough both at 200 hPa and 500 hPa levels. In addition, the cold air mass at 850 hPa was embedded into warm core structure totally, but the 1000–500 hPa thickness field formed a ‘tongue-like’structure (Figs.3d and 4d) near the low-pressure center. This ‘tongue-like’ geopotential thickness field suggested the incomplete vanishing of warm core, as the 500–1000 hPa thickness value is used to define ‘bulk’ air mass mean temperature.

and 1000–500 hPathick ness (blue contour, un it: gpm) at (a) 1800 UTC 28, (b) 0600 UTC 29, (c) 0000 UTC 30, and (d) 1200 UTC 30 October2012. The cyclone centers are indicated with symbols ‘’ for hurricane or ‘L’ for extratropicallow pressure system.

Fig.4 500 hPa geopotential height (black contour, unit: gpm), 850 hPa potential temperature (shaded, lower than 291 K, unit: K), 200 hPa minus 850 hPa wind shear (arrows, unit: m s-1) and 600–400 hPa mean ascent (red contour, beginning from -0.5 Pa s-1, unit: Pa s-1) at (a) 1800 UTC 28, (b) 0600 UTC 29, (c) 0000 UTC 30, and (d) 1200 UTC 30 October 2012.The cyclone centers are indicated with symbols ‘’ for hurricane or ‘L’ for extratropical low pressure system.

Fig.5 (a) Surface observations at 1500 UTC 29 October and GOES-E visible satellite image at 1445 UTC 29 October overlapped by fronts and isobars (4 hPa interval) analysis that provided by NHC at 1500 UTC 29 October. (b) Surface observations at 2100 UTC 29 October and GOES-E visible satellite image at 2045 UTC 29 October overlapped by fronts and isobars (4 hPa interval) analysis that provided by NHC at 2100 UTC 29 October (from Fig.22 and Fig.23 of Blake et al., 2013).

4.2 Intensification During Sandy’s ET Process

From both observational data and reanalysis data we can find that Hurricane Sandy experienced the ET process from late 28 October to late 29 October, and it became an extratropical cyclone prior to its landfall along the New Jersey shoreline around 2330 UTC October 29. During that period, Sandy reintensified from a weak category-1 hurricane with a wind speed around 35 m s-1to a category-2 hurricane with a wind speed around 45 m s-1near surface. Moreover, the minimum central sea level pressure even dropped down to 940 hPa at 1800 UTC 29 October simultaneously. This sharply increased wind speed accompanied by a rapidly decreased sea level pressure caused a devastating storm surge–the averaged water level rise at New York City came up to 3.96 m above normal tide level, and an easterly fetch that affected the entire eastern coast of United States (Blake et al., 2013).In order to investigate this extraordinary reintensification deeply, we will use more observational data as well as the WRF simulation results to analyze this process.

We now analyse the vertical sections which respectively crossing the Sandy center at four different times (1800 UTC 28, 0900 UTC 29, 2100 UTC 29 and 1200 UTC 30 October) to illustrate the evolutionary processes of Sandy’s secondary peak from the perspectives of geopotential height, potential temperature, three dimensional wind speed, as well as potential vorticity. Each vertical section is designed in 20? longitude by 20? latitude and center crossed Sandy’s inner core (Fig.6).

Fig.6 Sandy track simulated by WRF from 1200 UTC 28 to 1200 UTC 30 October 2012, and the orientations of cross sections. Each section is 20? longitude by 20? latitude, centering with Sandy inner core. Number 1–4 indicate the time at 1800 UTC 28, 0900 UTC 29, 2100 UTC 29 and 1200 UTC 31 October 2012, respectively.

TC is a rapidly rotating vortex system characterized by a low-pressure center, surrounded by great pressure gradient confined to the inner core and decreasing with height (Frank, 1977). By 1800 UTC 28 October, the deep negative geopotential height anomaly could reach to 300 hPa, while the 0 isoline had a great northwestward tilting over 400 hPa, resulting from the approaching of upstream upper-level trough. The whole system was really deep, with obvious warm core inside (Figs.7a, 7d). From both geopotential height anomaly and wind speed fields, we could find a well-defined eyewall structure, while deep convection was only limited in a narrow band located in the west and north of Sandy (Figs.7b, 7e). Additionally, a upper-level jet stream stayed in the south of Sandy near 200 hPa height, and the horizontal wind speed in the south was apparently greater than that of the north of Sandy center. The high PV value was located near the kernel structure with high PV gradient confined to the inner core (Figs.7c, 7f), just like a mature TC potential vorticity distribution (Kofron et al., 2010). Hurricane Sandy had a typical mature TC characteristic at this time.

Early on 29 October, Sandy made its forth turning, and advanced to the northeastern coastline of the United States. At 0900 UTC 29 October, due to the interaction with upper-level trough, the negative geopotential height anomaly was not as robust as 15 hours ago, and the system got a prominent trend to westward tilt above 600 hPa (Fig.8a). The upward motion in the western side of Sandy was strengthened, which was maybe caused by the increased upper-level divergence (Figs.3b, 8b). Despite the upward motion, the wind speed also had a significant increase on both of the northern and southern sides from bottom to over 400 hPa (Figs.8b, 8e). The weakened vertical wind shear further contributed to Sandy’s intensification (Fig.4b). What is more, besides the inner core high PV axis, a upper-level high PV was center located in west of low-level PV center (Fig.8c), which would strength the vortex through a series of adjustments (Hoskins et al., 1985). After that, Sandy moved across the main axis of Gulf Stream near 1200 UTC 29 October (Fig.9a), where there is a great enhancement in SST. Higher surface temperature is theoretically favorable to supply more energy to TC (Evans, 1993; Schade, 2000). The sounding profile observed at the station 72501 (40.86?N, 72.86?W, marked in Fig.2a) at 1200 UTC 29 October is shown in Fig.9b. As the dew-point deficit was almost zero under 500 hPa, it suggested that environmental condition was really humid due to the high SST. Additionally, the hodograph and wind barbs suggested that vertical wind shear was also really small both in magnitude and direction. The increased upper-level divergence and reduced vertical wind shear together with the higher SST caused Sandy’s reintensification.

Sandy reached its second peak at 1800 UTC 29 Octo-ber. By 2100 UTC 29 October, Sandy still maintained its strong westward tilting structure, but cold air mass had occupied the south and west of Sandy, and invaded the warm core from upper-level to bottom in the east of inner center, i.e., cold continental air was encircled inside Sandy cyclonically (Figs.10a, 10d). Although both vertical and horizontal windswere reinforced remarkably, Sandy began to lose its symmetric wind field gradually (Figs.10b, 10e). At the same time, potential vorticity displayed an interesting vertical distribution: although the major high PV axis slightly tilted to the west from top to bottom, another 500 hPa high PV center traveled ahead of the low level center (Figs.10c, 10f). This unusual PV configuration indicated the maturity of Sandy’s secondary peak as well as the upcoming dissipation (Hoskins et al., 1985).

After Sandy was reintensified, it moved over much colder water area, in which SST dropped from 25℃ to 17℃. Sandy made its landfall to the United States around 2330 UTC 29 October. Surface friction and cold air mass in conjunction with lowerSST weakened this system and hastened its loss of tropical structure. By 1200 UTC 30 October, Sandy had been completely integrated intopreexisting mid-latitude low-pressure system. The inner core of Sandy no longermaintained its structure as tightly as previous, though the central pressure was still lower than the surrounding and pressure gradient weakened significantly (Figs.11a, 11d). Due to the invasion of cold conti-nental air mass, the warm core structure almost vanished below 700 hPa but was still activated above 500 hPa. The wind field weakened and became asymmetric (Figs.11b, 11e). The potential vorticity near the inner core also reduced dramatically and became less concentrated (Fig.11c), which suggests that Sandy became an extratropical cyclone and dissipated much.

Fig.7 Vertical cross section analyses at 1800 UTC 28 October 2012. (a)–(c) W1–E1 and (d)–(f) N1–S1 orientations are shown in Fig.6. (a), (d): geopotential height anomaly (contours, zero line is marked by blue color; unit: gpm) and potential temperature anomaly (shaded; unit: K), each mean field is determined from the longitude/latitude that shown above; (b),(e): horizontal windspeed (contours, red contours mark the wind speed greater than and upward mo-tion (shaded; unit: (c), (f): potential vorticity (unit: PVU). The cyclone center is denoted by the symbol ‘’.

Fig.8 As in Fig.7, but at 0900 UTC 29 October 2012. (a)–(c) W2–E2 and (d)–(f) S2–N2 orientations are shown in Fig.6.The cyclone center is denoted by the symbol ‘’.

Fig.9 (a) Threedays (28, 29, 30 October) mean sea surface temperature (SST, unit: ℃). The cyclone centers are marked bythe symbol of ‘’ and ‘’. (b) Sounding data and hodograph of the station 72501 (Fig.2a) at 1200 UTC 29 October 2012. Solid line indicates air temperature (unit: ℃), and dashed line indicates dew-point temperature (unit: ℃).

5 Conclusions

In this paper, we made a systematic investigation regarding the ET process of Hurricane Sandy observationally and numerically. WRF3.3.1 model successfully reproduces the ET of Sandy that began on 29 October 2012 when Sandy interacted with a pre-existing upper-level negative tilt trough. The negative tilt trough enhanced upper-level divergence and decreased vertical wind shear significantly, which further contributed to deep convection. In addition to higher SST near Gulf Stream, Hurricane Sandy began to reintensified results from the cyclonic penetration of cold continental air that from back of upper-level trough and wrapped into the Sandy’s warm core. This strong continental cold air mass further caught up the warm front which located at the east of Sandy’s center, making Sandy get the characteristic of warm seclusion cyclone. This warm seclusion structure during ET process also provided a positive environmental condition for Sandy’s reintensification. Sandy reached its second peak around 1800 UTC 29 October with 940 hPa central sea level pressure and 45 m s-1wind speed. Late on 29 October, the cold air mass invaded the warm core from upper-level to low-level in the east of Sandy. Surface fric-tion, lower SST and involvement of cold continental air mass accelerated Sandy’s dissipation. Early on 30 October, both the decreased inner core pressure and PV gradient associated with asymmetric wind field suggested Sandy’s loss of tropical structure. At last, Sandy completely fused with a small low-pressure system from midlatitude negative tilt trough and became an extratropical cyclone.

Fig.10 As in Fig.7, but at 2100 UTC 29 October 2012. (a)–(c) W3–E3 and (d)–(f) S3–N3 orientationsare shown in Fig.6.The cyclone center is denoted by the symbol ’.

Fig.11 As in Fig.7, but at 1200 UTC 30 October 2012. (a)–(c) W4–E4 and (d)–(f) S4–N4 orientations are shown in Fig.6. The cyclone center is denoted by the symbol ‘L’.

Acknowledgements

References

Bergeron, T., 1954. Review of modern meteorology – 12. The problem of tropical hurricanes. Quarterly Journal of the Royal Meteorological Society, 80: 131-164.

Betts, A. K., 1986. A new convective adjustment scheme. Part I: Observational and theoretical basis. Quarterly Journal of the Royal Meteorological Society, 112: 677-691.

Betts, A. K., and Miller, M. J., 1986. A new convective adjustment scheme. Part II: Single column tests using GATE wave, BOMEX, and arctic air-mass data sets. Quarterly Journal of the Royal Meteorological Society, 112: 693-709.

Blake, E. S., Kimberlain, T. B., Berg, R. J., Cangialosi, J. P., and Beven II, J. L., 2013. Tropical Cyclone Report Hurricane Sandy (AL182012) 22–29 October 2012. National Hurricane Center.

Browning, K. A., 2004. The sting at the end of the tail: Damaging winds associated with extratropical cyclones. Quarterly Journal of the Royal Meteorological Society, 130: 375-399.

Browning, K. A., Vaughan, G., and Panagi, P., 1998. Analysis of an ex-tropical cyclone after reintensifying as a warm-core extratropical cyclone. Quarterly Journal of the Royal Meteorological Society, 124: 2329-2356.

Carr, F. H., and Bosart, L. F., 1978. A diagnostic evaluation of rainfall predictability for tropical storm Agnes, June 1972. Monthly Weather Review, 106: 363-374.

科技的發(fā)展使得現(xiàn)階段我國(guó)農(nóng)業(yè)生產(chǎn)得到不斷的提高，但是農(nóng)業(yè)由于受到氣象因素的影響較大，因此良好的氣象監(jiān)測(cè)是實(shí)現(xiàn)農(nóng)業(yè)進(jìn)一步發(fā)展的基礎(chǔ)。我國(guó)現(xiàn)階段自動(dòng)氣象站的使用已經(jīng)得到普及，基于自動(dòng)氣象站的各項(xiàng)技術(shù)也日趨成熟，得到大范圍的使用。

Chen, F., and Dudhia, J., 2001. Coupling an advanced landsurface/hydrology model with the Penn State/NCAR MM5 modeling system. Part I: Model description and implementation. Monthly Weather Review, 129: 569-585.

DiMego, G. J., and Bosart, L. F., 1982. The transformation of Tropical Storm Agnes into an extratropical cyclone. Part I: The observed fields and vertical motion computations. Monthly Weather Review, 110: 385-411.

Dudhia, J., 1989. Numerical study of convection observed during the winter monsoon experiment using a mesoscale twodimensional model. Journal of the Atmospheric Sciences, 46: 3077-3107.

Evans, C., and Hart, R. E., 2008. Analysis of the wind field evolution associated with the extratropical transition of Bonnie (1998). Monthly Weather Review, 136: 2047-2065.

Evans, J. L., 1993. Sensitivity of tropical cyclones to sea surface temperature. Journal of Climate, 6: 1133-1140.

Evans, J. L., and Hart, R. E., 2003. Objective indicators of the life cycle evolution of extratropical transition for Atlantic tropical cyclones. Monthly Weather Review, 131: 909-925.

Foley, G. R., and Hanstrum, B. N., 1994. The capture of tropical cyclones by cold fronts off the west coast of Australia. Weather and Forecasting, 9: 577-592.

Frank, W. M., 1977. The structure and energetics of the tropical cyclone I. Storm structure. Monthly Weather Review, 105: 1119-1135.

Galarneau, T. J., Davis, C. A., and Shapiro, M. A., 2013. Intensification of Hurricane Sandy (2012) through extratropical warm core seclusion. Monthly Weather Review, 141: 4296-4321.

Hart, R. E., and Evans, J. L., 2001. A climatology of the extratropical transition of Atlantic tropical cyclones. Journal of Climate, 14: 546-564.

Hong, S.-Y., Noh, Y., and Dudhia, J., 2006. A new vertical diffusion package with an explicit treatment of entrainment processes. Monthly Weather Review, 134: 2318-2341.

Hoskins, B. J., McIntyre, M. E., and Robertson, A. W., 1985. On the use and significance of isentropic potential vorticity maps. Quarterly Journal of the Royal Meteorological Society, 111: 877-946.

Janjic, Z. I., 1994. The step-mountain eta coordinate model: Further developments of the convection, viscous sublayer, and turbulence closure schemes. Monthly Weather Review, 122: 927-945.

Jones, S. C., Harr, P. A., Abraham, J., Bosart, L. F., Bowyer, P. J., Evans, J. L., Hanley, D. E., Hanstrum, B. N., Hart, R. E., Lalaurette, F., Sinclair, M. R., Smith, R. K., and Thorncroft, C., 2003. The extratropical transition of tropical cyclones: Forecast challenges, current understanding, and future directions. Weather and Forecasting, 18: 1052-1092.

Klein, P. M., Harr, P. A., and Elsberry, R. L., 2000. Extratropical transition of western North Pacific tropical cyclones: An overview and conceptual model of the transformation stage. Weather and Forecasting, 15: 373-396.

Kofron, D. E., Ritchie, E. A., and Tyo, J. S., 2010. Determination of a consistent time for the extratropical transition of tropical cyclones. Part II: Potential vorticity metrics. Monthly Weather Review, 138: 4344-4361.

Lin, Y.-L., Farley, R. D., and Orville, H. D., 1983. Bulk parameterization of the snow field in a cloud model. Journal of Climate and Applied Meteorology, 22: 1065-1092.

Michael, B., 2012. Hurricane Sandy Discussion NuhPaer 14 (Report). National Hurricane Center.

Mlawer, E. J., Taubman, S. J., Brown, P. D., Iacono, M. J., and Clough, S. A., 1997. Radiative transfer for inhomogeneous atmosphere: RRTM, a validated correlated-k model for the longwave. Journal of Geophysical Research, 102 (D14): 16663-16682.

Noh, Y., Cheon, W. G., Hong, S. Y., and Raasch, S., 2003. Improvement of the K-profile model for the planetary boundary layer based on large eddy simulation data. Boundary-Layer Meteorology, 107: 401-427.

Palmén, E., 1958. Vertical circulation and release of kinetic energy during the development of hurricane Hazel into an extratropical storm. Tellus, 10: 1-23.

Ritchie, E. A., and Elsberry, R. L., 2001. Simulations of the transformation stage of the extratropical transition of tropical cyclones. Monthly Weather Review, 129: 1462-1480.

Schade, L. R., 2000. Tropical cyclone intensity and sea surface temperature. Journal of the Atmospheric Sciences, 57: 3122-3130.

Sekioka, M., 1970. On the behaviour of cloud pattern as seen on satellite photographs in the transformation of a typhoon into an extratropical cyclone. Journal of the Meteorological Society of Japan, 48: 224-233.

Sekioka, M., 1972. Note on the extratropical transformation of a typhoon in relation with old outbreaks. Archiv für Meteorologie, Geophysik und Bioklimatologie, A21: 413-418.

Shapiro, M. A., and Keyser, D., 1990. Fronts, jet streams, and the tropopause. Extratropical Cyclones: The Erik Palmén Memorial Volume. Newton, C. W., and Holopaninen, E. O., eds., American Meteorological Society, 167-191.

Thorncroft, C. D., and Jones, S. C., 2000. The extratropical transition of Hurricanes Felix and Iris in 1995. Monthly Weather Review, 128: 947-972.

Wang, Q., Q. Li, and G. Fu, 2012: Determining the Extratropical Transition Onset and Completion Times of Typhoons Mindulle (2004) and Yagi (2006) Using Four Methods. Wea. Forecasting, 27, 1394–1412.

(Edited by Xie Jun)

(Received September 30, 2014; revised November 28, 2014; accepted April 13, 2015)

J. Ocean Univ. China (Oceanic and Coastal Sea Research)

DOI 10.1007/s11802-015-2770-2

ISSN 1672-5182, 2015 14 (5): 783-794

http://www.ouc.edu.cn/xbywb/

E-mail:xbywb@ouc.edu.cn

* Corresponding author. Tel: 0086-532-66781920

E-mail: fugang@ouc.edu.cn