1. Introduction
A close relationship has been established between the Atlantic tropical cyclone (TC) activity and sea surface temperature (SST) based on observations especially since the 1970s (Emanuel 2005; Webster et al. 2005; Emanuel 2008). A few studies have suggested that the upswing of TC activity in recent decades is accompanied with enhanced TC formation over the eastern tropical Atlantic (Holland 2007; Wu and Wang 2008; Wu et al. 2010). Over the longer period, however, the TC frequency records are subject to uncertainty. Whereas Vecchi and Knutson (2008) estimated that the undercount in the TC database was ¼ TC yr−1 in the 1950s and 1960s, Landsea (2007) argued that numerous TCs were missed in the existing Atlantic TC count database with 2.2 TCs yr−1 missed for the period 1900–65. In addition, Landsea et al. (2010) suggested that the previously documented increase in Atlantic total TC frequency is due primarily to an increase in short-lived (<2 days) TCs, but Villarini et al. (2011) suggested that the observed increasing trend in short-lived TCs is spurious because of changes in the observational system.
While some studies related the enhanced TC formation to the ongoing global warming (Emanuel 2005; Hoyos et al. 2006; Mann and Emanuel 2006; Trenberth and Shea 2006; Holland and Webster 2007; Mann et al. 2007a,b), other studies attributed it to the warm phase of the Atlantic multidecadal oscillation (AMO), which is usually associated with heightened TC activity in the North Atlantic basin (Delworth and Mann 2000; Enfield et al. 2001; Goldenberg et al. 2001; Landsea 2005; Pielke et al. 2005; Trenberth and Shea 2006; Knight et al. 2006; Kossin and Vimont 2007; Enfield and Cid-Serrano 2010). So far the attribution of the enhanced Atlantic TC formation in the recent decades still remains as a scientific issue.
The SST variability in the North Atlantic Ocean on time scales longer than interannual is mainly characterized by an AMO-scale variation and a secular trend (Trenberth and Shea 2006; Ting et al. 2009; Wang et al. 2012). The AMO variability in the North Atlantic SST shows a quasi-cycle of 60–100 yr (Gray et al. 2004). With a warm period of 1930–60 and a cold period of 1970–90, the AMO is currently in its warm phase (Curry 2008; Ting et al. 2009; Enfield and Cid-Serrano 2010). It is quite obvious that the impact of global warming on the Atlantic TC activity is entangled by the AMO variability. Direct extraction of this impact has been difficult because of the uncertainty and limited sample size of TC formation in the Atlantic basin. An alternate approach that is usually used to qualitatively understand the influence of global warming on TC activity is to identify how the associated large-scale environment changes since it is well known that the climatologic aspects of TC formation are closely related to a few large-scale environmental parameters including SST, low-level vorticity, vertical wind shear, and midlevel relative humidity (Gray 1968, 1975; Knaff 1997; Shapiro and Goldenberg 1998; Goldenberg et al. 2001; Bell and Chelliah 2006; Wu et al. 2010).
Following Gray (1968, 1975), Emanuel and Nolan (2004) quantified the influences of these large-scale environmental parameters on TC formation by defining an empirical genesis potential index (GPI), which includes the effects of 850-hPa absolute vorticity, 700-hPa relative humidity, vertical wind shear between 850 and 200 hPa, and the maximum potential intensity (MPI). The MPI is mainly a function of SST and the outflow temperature (Emanuel 1986). Camargo et al. (2007b) showed that the TC activity difference between El Niño and La Niña years can be demonstrated in the GPI. Moreover, the GPI was also used to explore the effect of large-scale environment on TC formation in numerical models (Nolan et al. 2006; Camargo et al. 2007a,b; Murakami and Wang 2010; Jiang et al. 2012). These studies suggest that the GPI is useful to understand TC formation due to changes in large-scale environmental conditions.
In this study, an attempt is made to distinguish the influences of the AMO and global warming on TC formation in the North Atlantic basin. For this purpose the National Oceanic and Atmospheric Administration (NOAA) Earth System Research Laboratory (ESRL) Twentieth Century Reanalysis (20CR) dataset is used. The three-dimensional structure of the time-evolving state in the dataset is numerically produced by using only observed SST, sea ice, and surface pressure (Compo et al. 2006, 2008). Despite a relative lack of observational constraints above the surface, the resultant time series of the atmospheric state may be relatively free of strong biases in the dataset due to relatively homogeneous observations of pressure, sea ice, and sea surface temperature through the whole period (Emanuel 2010). Following the introduction of the datasets used in this study in section 2, we demonstrate in section 3 that variations in the Atlantic TC formation with time scales longer than interannual can be fairly well represented by the GPI when we calculate it using the 20CR dataset. In sections 4 and 5, two leading patterns in TC formation that are associated with AMO and global warming are first extracted through the empirical orthogonal function (EOF) analysis of the GPI and then the key features of large-scale parameters associated with the two patterns are identified through regression analysis. A summary is finally presented in section 6.
2. Data
The Atlantic hurricane track data are from the National Hurricane Center (NHC). TC frequency records are relatively reliable from 1945 onward because of the implementation of aircraft reconnaissance and the relatively dense ship traffic (Vecchi and Knutson 2008). For comparison, the GPI defined by Emanuel and Nolan (2004) is calculated based on several reanalysis datasets: the NOAA ESRL Twentieth Century Reanalysis, version 2 (20CRv2; 2° latitude × 2° longitude with 24 vertical levels; Compo et al. 2011), the National Centers for Environmental Prediction–National Center for Atmospheric Research reanalysis (NCEP–NCAR; 2.5° latitude × 2.5° longitude with 17 vertical levels; Kalnay et al. 1996), the 40-yr European Centre for Medium-Range Weather Forecasting (ECMWF) Reanalysis (ERA-40; 2.5° latitude × 2.5° longitude with 23 vertical levels; Uppala et al. 2005) and the ECMWF Interim Re-Analysis (ERA-Interim; 1.5° latitude × 1.5° longitude with 37 vertical levels; Simmons et al. 2007), the Japanese 25-yr Reanalysis (JRA-25; 1.25° latitude × 1.25° longitude with 23 vertical levels; Onogi et al. 2005), and the National Aeronautics and Space Administration (NASA) Modern-Era Reanalysis for Research and Applications (MERRA; ½° latitude × ⅔° longitude with 42 vertical levels; Rienecker et al. 2011). Since the 20CR and NCEP–NCAR data are available prior to 2011 and after 1948, our analysis in this study covers only the peak hurricane season (August–October) during the period 1948–2010. A five-point smoother is applied to reduce interannual variability.
The monthly anomalies of the combined mean global land and ocean temperature and the time series of the AMO index are also used from the National Climate Data Center (http://www.ncdc.noaa.gov/cmb-faq/anomalies.html#anomalies) and the ESRL (http://www.cdc.noaa.gov/correlation/amon.us.long.data; Enfield et al. 2001), respectively. The SST data are from the NOAA extended reconstructed SST (ERSST version 3) data with 2° latitude by longitude 2° resolution (Smith et al. 2008).
3. Relationship between TC formation and GPI
Although efforts to improve the GPI have been made in recent years (Emanuel 2010; Murakami and Wang 2010; Tippett et al. 2011; McGauley and Nolan 2011), Bruyere et al. (2012) and Menkes et al. (2012) recently evaluated TC genesis indices on seasonal and interannual time scales. They found that all TC genesis indices struggle with reproducing interannual frequency variability and trend. Menkes et al. (2012) suggested that differences between TC genesis indices are large and vary depending on the regions and on the time scales considered. In what follows, we show that long-term changes in the Atlantic TC formation are fairly well represented by the GPI proposed by Emanuel and Nolan (2004) when we calculate it using the 20CR dataset.
Figure 1 shows the spatial distribution of the GPI derived from the monthly 20CR dataset, comparing with the observed formation frequency in grid boxes of 2.5° latitudes by 2.5° longitudes and the GPI calculated from the monthly NCEP–NCAR, MERRA, ERA-Interim, JRA-25, and ERA-40 datasets. The TC formation frequency indicates how many TCs (maximum sustained winds exceed 17 m s−1) form in a specific grid box of 2.5° latitude by 2.5° longitude per year. The TC formation frequency and GPI are averaged in the peak hurricane season (August–October) over the period 1979–2010 except for the ERA-40 dataset (1979–2001). For comparison, the GPI calculated with varying horizontal resolutions is interpolated to a uniform resolution of 2.5° latitude by 2.5° longitude. In the observation (Fig. 1a), the relatively large formation frequency is observed roughly in two zones. One is from the eastern tropical Atlantic to the Caribbean Sea between 5° and 20°N and the other extends from the Gulf of Mexico to the western Atlantic, in general between 20° and 30°N. This feature can be found in the GPI from all of the reanalysis datasets although the formation is overestimated in the subtropics and near the west coast of Africa due to the lack of the land effect in the GPI calculation. Figure 1 suggests that the GPI can capture the main features in the mean spatial distribution of the observed TC formation, in particular for the 20CR (Fig. 1b), MERRA (Fig. 1d), and ERA-Interim (Fig. 1e) datasets, while the GPI from the NCEP–NCAR (Fig. 1c), JRA-25 (Fig. 1f), and ERA-40 (Fig. 1g) datasets underestimates the TC formation in the central tropical Atlantic region (30°–60°W). Figure 1 also suggests that the mean spatial distribution of the GPI derived from the 20CR reanalysis is fairly well comparable to the observation and the MERRA and ERA-Interim datasets, although the two datasets currently represent the state-of-the-art reanalysis with a relatively high horizontal resolution (Simmons et al. 2007; Rienecker et al. 2011).
(a) August–October mean spatial distribution of tropical cyclone formation frequency and the GPI derived from (b) the Twentieth Century Reanalysis (20CR), (c) NCEP–NCAR, (d) MERRA, (e) ERA-Interim, (f) JRA-25, and (g) ERA-40 over the period 1979–2010 except for the ERA-40 dataset (1979–2001). The unit of tropical cyclone formation frequency is yr−1. Note that the color scales are different in different panels.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
One may notice the differences of the GPI magnitude in the different datasets in Fig. 1. Our calculation reveals that the magnitude differences result mainly from the differences in the 600-hPa relative humidity in these datasets (not shown). Usually higher relative humidity leads to larger magnitudes in the GPI.
Following Bruyere et al. (2012) and Menkes et al. (2012), we further examine the temporal changes in the GPI-derived TC formation frequency against the observed TC frequency in the peak hurricane season, with a focus on the multidecadal time scale (Fig. 2). The GPI-derived formation frequency is obtained by averaging the GPI over the area of 5°–35°N, 10°–100°W. For comparison, all of the time series are standardized with a 5-yr running mean. In Fig. 2a, we first show the GPI time series derived from the three high-resolution reanalysis datasets (MERRA, JRA-25, and ERA-Interim). Despite differences on the decadal time scale, the time series from the three datasets all present an increasing trend over the past three decades, which is in general consistent with the observed increase in the total TC and long-lived (≥2 days) TC formation frequency. As shown in Fig. 2b, we can see that the GPI time series from the 20CR dataset is better than those from the NECP–NCAR and the ERA-40 datasets since the overall changes over the period 1950–2008 can be clearly revealed in the 20CR data despite a discrepancy in early 1990s. The GPI time series calculated from the 20CR dataset shows highest correlation (~0.7) with the observed TC formation frequency over the period 1950–2008 in the three longer datasets, significant at the 95% confidence level based on the effective degree of freedom. Note that the time series from the NCEP–NCAR reanalysis shows a decreasing trend in the GPI before the mid-1980s, which results mainly from the decreasing trend in tropical middle and upper tropospheric specific humidity, as demonstrated by Dessler and Davis (2010).
(a) Standardized time series of the 5-yr running mean frequency of tropical cyclones (black), long-lived (≥2 days) tropical cyclones (dashed black), and the corresponding GPI from JRA-25 (red), MERRA (blue), and ERA-Interim (green) from 1979 to 2010, and (b) time series of the annual frequency of tropical cyclones (black), long-lived (≥2 days) tropical cyclones (dashed black) and the GPI from 20CR (red), NCEP–NCAR reanalysis (blue), and ERA-40 (green) from 1948 to 2010 (1958–2001 for the ERA-40 dataset) in August–October. The GPI is averaged obtained over the area of 5°–35°N, 10°–100°W.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
As shown by Holland (2007), Wu and Wang (2008), and Wu et al. (2010), the increasing TC activity over the recent two decades was characterized by the enhanced TC formation over the eastern tropical Atlantic. This feature can be seen in the epochal difference of the GPI derived from the 20CR between the periods 1971–90 and 1991–2010 (Fig. 3). Figure 3 suggests that the GPI-derived formation frequency fairly well reproduces the enhancement of the observed TC formation in the eastern tropical North Atlantic. The significant increasing formation extends from the African coast to about 60°W although Wu et al. (2010) found that the observed one occurred mainly east of 45°W.
The difference in August–October mean GPI derived from 20CR between the periods 1971–90 and 1991–2010. The rectangle denotes the region 10°–20°N, 15°–45°W with enhanced TC formation in the observation. The contour interval is 0.5 with the shaded areas indicating significant at the 95% confidence level.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
Landsea et al. (2010) argued that the previously documented increase in Atlantic total TC frequency is due primarily to an increase in short-lived (<2 days) TCs since the late 1800s. Figure 4 shows the 5-yr running mean of annual frequency of all TCs (maximum sustained winds exceed 17 m s−1) and long-lived (≥2 days) TCs during the period 1950–2008. In agreement with Landsea et al. (2010), the increasing trend in the annual TC counts is remarkably reduced when short-lived TCs are removed from the dataset (Fig. 4a). However, removal of the short-lived TCs primarily affects the region west of 45°W and in fact no overall trend can be seen for this region if the short-lived TCs are removed from the time series (Fig. 4b). On the other hand, the removal of short-lived TCs has little effect on the increasing trend in the eastern tropical Atlantic (Fig. 4c). This is reasonable since TCs that form over the east tropical Atlantic can have longer duration and more potential to develop into intense hurricanes (Wu and Wang 2008). Close inspection indicates that the increasing trend in the east tropical Atlantic can be identified since the 1970s (Fig. 4c), while the increasing trend for all TCs started from the 1990s (Fig. 4a).
The 5-yr running mean of the annual frequency of all tropical cyclones (solid) and long-lived (≥2 days) tropical cyclones (dashed) superimposed on the climatological mean value that formed in (a) 10°–100°W, (b) 45°–100°W, and (c) 10°–45°W in August–October during 1948–2010, respectively.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
4. Leading patterns in TC formation change
In the last section, we have shown that the GPI calculated from the 20CR dataset is consistent with the observed TC formation in terms of the mean spatial distribution over the past three decades, the GPI-derived formation frequency, and the epochal difference of TC formation in the east tropical Atlantic. This encourages us to further understand the changes in TC formation with time scales longer than interannual through analysis of the GPI derived from the 20CR dataset. To obtain the leading patterns in the North Atlantic TC formation, the conventional EOF analysis technique is applied to the GPI field derived from the 20CR over the period 1948–2010. Note that a similar analysis to the observed TC frequency is nearly impossible due to the limited number of samples each year.
Figure 5 shows the first two leading patterns, which explain 37.7% and 21% of the total variance, respectively. The spatial distribution of the first pattern shows positive loadings in the tropical Atlantic, the Caribbean Sea, and the Gulf of Mexico (Fig. 5a), suggesting the enhanced TC formation during the positive phase of this pattern. Negative loadings are found over the areas north of 25°N and south of 10°N. The time series of the first principal component (PC1) shows a transition from the positive phase to the negative phase from the early 1950s to the mid-1980s, becoming positive over the past two decades (Fig. 5b). This feature is similar to the AMO variability in the basinwide North Atlantic SST (Curry 2008; Ting et al. 2009; Enfield and Cid-Serrano 2010). Goldenberg et al. (2001) suggested that the recent enhancement of the Atlantic TC formation was related to the positive phase of the AMO since the 1990s. Note that PC1 did not increase until the mid-1980s, while the AMO index turned to increase in the early 1970s. One possible reason is that the AMO index is calculated only as the detrended North Atlantic SST anomalies over the region of 0°–60°N and from the east coast of the Americas to 0° longitude (Wang et al. 2012), whereas the GPI also include the effects of vertical wind shear, midlevel relative humidity, and low-level vorticity (Emanuel and Nolan 2004).
(a) The first spatial pattern of the GPI derived from 20CR with the contour intervals of 0.02 and (b) the associated time series (black) from 1948 to 2010 in comparison with the AMO index (red). (c) The second spatial pattern of the GPI derived from 20CR with contour intervals of 0.02 and (d) the associated time series (black) from 1948 to 2010 in comparison with the time series of the combined mean global land and ocean temperature anomaly (red). Standardization is performed for the two time series.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
Figure 6a displays the correlation distribution between the PC1 and the global SST, indicating an interhemispheric dipole with warm SST in the North Atlantic and the Caribbean Sea and the Gulf of Mexico, as well as cold one in the South Atlantic. Previous studies have suggested that this SST pattern is associated with the AMO (Knight et al. 2005; Sutton and Hodson 2005; Trenberth and Shea 2006; Zhang and Delworth 2006; Baines and Folland 2007; Hodson et al. 2009; Ting et al. 2009; Wang et al. 2012). While the negative correlations can be found in the tropical eastern Pacific and high-latitude South Pacific, the positive centers can be found in the midlatitude Pacific in the Southern and Northern Hemispheres. The first pattern is thus called the AMO pattern of TC formation in this study.
Correlations of SST with (a) PC1 and (b) PC2 in August–October during the period 1948–2010. The contour intervals are 0.2. The shaded areas indicate significant at the 95% confidence level based on the effective degree of freedom. A 5-yr running mean is performed before the correlation calculation.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
The main feature of the second EOF spatial pattern is the remarkable positive loadings in the east tropical Atlantic (5°–20°N, 15°–40°W), suggesting enhanced TC formation during the positive phase (Fig. 5c). Negative loadings are located mainly from the southeast coast of the United States extending southward to the Caribbean Sea. Figure 5d shows the corresponding time series or the second principal component (PC2), indicating a significant increasing trend since the 1960s. The second pattern was in the negative phase during 1950–86 and then changed into the positive phase. In addition to the AMO pattern, this suggests that the second pattern also contributes to the increasing trend of TC formation in the east North Atlantic over the recent decades. Further analysis indicates that the pattern is highly correlated (r = 0.93) with monthly anomalies of the combined mean global land and ocean temperature (Fig. 5b), i keeping with the definition of global warming (Solomon et al. 2007). The correlation distribution of the PC2 with global SST shows a global-scale warming with remarkable warming in the Indian Ocean, western tropical Pacific, and tropical Atlantic in the positive phase of PC2 since the 1980s (Fig. 6b). Based on these features, the second pattern is called the global warming pattern in this study.
The 5-yr running mean of August–October frequency of all tropical cyclones (solid) and long-lived (≥2 days) tropical cyclones (dashed dotted), and the regressed one with PC1 and PC2 (short dashed) during the period 1948–2010.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
5. Large-scale factors associated with TC formation changes
Based on the above analysis, understanding of the TC formation changes associated with the AMO and global warming patterns can be improved by examining the contributions of large-scale factors, namely low-level vorticity, vertical wind shear, midlevel relative humidity, and MPI (Emanuel and Nolan 2004). To quantify the contribution of low-level vorticity, we first replace it with the climatological mean vorticity in the EOF analysis and find that the resulting spatial patterns and time series are nearly identical to those shown in Fig. 5 (figures not shown), suggesting that relative vorticity plays a negligible role in the TC formation changes with time scales longer than interannual. This agrees with Bruyere et al. (2012) that a revised index comprising only vertical wind shear and MPI has a significant skill at reproducing interannual variations and trends of the North Atlantic TC formation.
The correlation of PC1 and PC2 with the other parameters is conducted to identify their contributions to the AMO and global warming patterns. For the AMO pattern (Fig. 8), the negative correlations with vertical wind shear extend from the Africa to the Gulf of Mexico, covering most of the Atlantic TC activity region. This means that the warm AMO phase is associated with a basinwide decrease in vertical wind shear and thus enhanced TC formation, in agreement with Goldenberg et al. (2001). Numerical simulation in Zhang and Delworth (2006) showed reductions in vertical wind shear over the main development region (MDR) during the positive AMO phase. Kossin and Vimont (2007) and Wang et al. (2008) further suggested that vertical wind shear is affected by the AMO through the Atlantic meridional mode (AMM) and the Atlantic warm pool (AWP). An anomalously large AWP reduces both the lower tropospheric easterly winds and the upper tropospheric westerly winds, resulting in a reduction in vertical wind shear. Note that the spatial pattern of vertical wind shear is similar to the one regressed with dust aerosol in Wang et al. (2012). They argued that dust variability changes the meridional air temperature gradient through the dust-related shortwave and longwave radiative heating, altering the strength of the zonal winds or the easterly jet through the thermal wind balance and thus the vertical wind shear. There are significant positive correlations with the midlevel relative humidity in the eastern Atlantic and with MPI in the central tropical Atlantic, the Caribbean Sea, and the Gulf of Mexico, suggesting that the combined effect of increased midlevel relative humidity and MPI also favors the basinwide enhancement of TC formation in the AMO warm phase (Figs. 8b,c).
Correlations of PC1 with (a) vertical wind shear, (b) 600-hPa relative humidity, and (c) MPI in August–October during the period 1948–2010. The rectangle denotes the region 10°–20°N, 15°–45°W with enhanced TC formation in the observation. The contour intervals are 0.3, 0.2, and 0.2 in (a),(b), and (c), respectively. The shaded areas indicate significant at the 95% confidence level based on the effective degree of freedom. A 5-yr running mean is performed before the correlation calculation.
Citation: Journal of Climate 26, 22; 10.1175/JCLI-D-13-00056.1
Based on numerical experiments, Hagos and Cook (2008) and Wu et al. (2010) suggested that the SST warming in the tropical Atlantic induces anomalous low-level cyclones off the African coast, increasing middle-level relative humidity in the eastern tropical Atlantic. The positive correlations with the midlevel relative humidity in Fig. 8b support these numerical simulations. In Fig. 8c, the positive correlations with MPI over the central tropical Atlantic, the Caribbean Sea, and the Gulf of Mexico are associated with the positive SST loadings in the regions, as shown in Fig. 6a.
For the global warming pattern (Fig. 9), there are significant positive correlations with vertical wind shear and negative correlations with relative humidity and MPI in the west part of the basin. All of these features suggest that global warming suppress TC formation in the west part of the basin, especially from the southeast coast of the United States extending southward to the Caribbean Sea. Vecchi and Soden (2007) examined changes in vertical wind shear in the climate models of the Intergovernmental Panel on Climate Change Fourth Assessment Report under the emission scenario A1B and found a prominent increase in vertical wind shear over the tropical Atlantic. Figure 9a shows that the enhanced vertical wind shear associated with the global warming pattern occurs mainly in the Caribbean Sea and the Gulf of Mexico.
Figures 9b and 9c suggest that global warming enhances the TC formation in the central tropical Atlantic through increasing relative humidity and MPI, but the effect is moderate (Fig. 5c). The remarkable increase in Fig. 5c occurs in the east tropical Atlantic (5°–20°N, 15°–40°W), which results mainly from the increasing MPI from the central Atlantic to the eastern Atlantic. In other words, our analysis suggested that global warming leads to the enhancement of TC formation in the east tropical Atlantic due to increasing MPI or SST.
Wu et al. (2010) suggested that the enhanced TC activity including basinwide increases in the average lifetime, annual frequency, proportion of intense hurricanes, and annual accumulated power dissipation index (PDI) resulted from the enhanced TC formation in the east tropical Atlantic. They argued that the Atlantic sea surface warming that occurred in recent decades might have allowed more TCs to form, to form earlier, and to take a longer track. This study confirms that the enhanced TC formation in the east tropical Atlantic is due to the local increase in MPI or SST, leading to the close relationship between the Atlantic SST and TC activity over the past 30 years (Emanuel 2005; Webster et al. 2005).
6. Summary
The heightened TC activity in recent decades is accompanied with enhanced TC formation over the eastern tropical Atlantic (Holland 2007; Wu and Wang 2008; Wu et al. 2010). Since many studies have shown that the TC activity on the time scales longer than interannual is associated with AMO variability (Delworth and Mann 2000; Enfield et al. 2001; Goldenberg et al. 2001; Landsea 2005) and global warming (Emanuel 2005; Hoyos et al. 2006; Mann and Emanuel 2006; Trenberth and Shea 2006; Holland and Webster 2007; Mann et al. 2007a,b), an attempt in this study is made to distinguish the influences of the AMO and global warming on TC formation in the North Atlantic basin. We first demonstrate that variations of the Atlantic TC formation with time scales longer than interannual can be fairly well represented by the GPI calculated from the 20CR dataset. Two distinctive climate change patterns in TC formation that are associated with AMO and global warming are revealed and the associated key factors associated with the two patterns are discussed.
The EOF analysis of the GPI field leads to two leading patterns in the climate change of TC formation on time scales longer than interannual. The first pattern is associated with AMO and its spatial pattern shows the basinwide enhancement of TC formation during the AMO positive phase. The second pattern is associated with global warming, showing enhanced TC formation in the east tropical Atlantic (5°–20°N, 15°–40°W) and reduced TC formation from the southeast coast of the United States extending southward to the Caribbean Sea. The two climate change patterns in TC formation can account for the observed variability of the TC formation over the period 1950–2008.
Further analysis indicates that relative vorticity plays a negligible role in the TC formation changes with time scales longer than interannual. The warm AMO phase is associated with a basinwide decrease in vertical wind shear and thus enhanced basinwide TC formation. The combined effect of increased midlevel relative humidity and MPI also favors the basinwide enhancement of TC formation in the AMO warm phase. For the global warming pattern, significant positive correlations with vertical wind shear and negative correlations with relative humidity and MPI in the west part of the basin suggest that global warming suppresses TC formation from the southeast coast of the United States extending southward to the Caribbean Sea, while the increasing MPI from the central Atlantic to the eastern Atlantic plays an important role in the enhancement of TC formation in the east tropical Atlantic. This study confirms that the enhanced TC formation in the east tropical Atlantic is associated with the local increase in MPI or SST, leading to the close relationship between the Atlantic SST and TC activity over the past 30 years (Emanuel 2005; Webster et al. 2005).
Acknowledgments
We thank Dr. Christopher W. Landsea for valuable comments and suggestions that led to improvements of the manuscript. This research was jointly supported by the Typhoon Research Project (2009CB421503) of the National Basic Research Program (the 973 Program) of China, the National Natural Science Foundation of China (NSFC Grant 41275093), the Social Commonweal Research Program of the Ministry of Science and Technology of the People's Republic of China (GYHY200806009), and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). Support for the Twentieth Century Reanalysis Project dataset is provided by the U.S. Department of Energy, Office of Science Innovative and Novel Computational Impact on Theory and Experiment (DOE INCITE) program, the Office of Biological and Environmental Research (BER), and by the National Oceanic and Atmospheric Administration Climate Program Office.
REFERENCES
Baines, P. G., and C. K. Folland, 2007: Evidence for a rapid global climate shift across the late 1960s. J. Climate, 20, 2721–2744.
Bell, G. D., and M. Chelliah, 2006: Leading tropical modes associated with interannual and multidecadal fluctuations in North Atlantic hurricane activity. J. Climate, 19, 590–612.
Bruyere, C. L., G. J. Holland, and E. Towler, 2012: Investigating the use of a genesis potential index for tropical cyclones in the North Atlantic basin. J. Climate, 25, 8611–8626.
Camargo, S. J., K. A. Emanuel, and A. H. Sobel, 2007a: Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis. J. Climate, 20, 4819–4834.
Camargo, S. J., A. H. Sobel, A. G. Barnston, and K. A. Emanuel, 2007b: Tropical cyclone genesis potential index in climate models. Tellus, 59A, 428–443.
Compo, G. P., J. S. Whitaker, and P. D. Sardeshmukh, 2006: Feasibility of a 100-year reanalysis using only surface pressure data. Bull. Amer. Meteor. Soc., 87, 175–190.
Compo, G. P., J. S. Whitaker, and P. D. Sardeshmukh, 2008: The 20th Century Reanalysis Project. Proc. Third WCRP Int. Conf. on Reanalysis, The University of Tokyo, Japan, WCRP. [Available online at http://wcrp.ipsl.jussieu.fr/Workshops/Reanalysis2008/Documents/V5-511_ea.pdf.]
Compo, G. P., and Coauthors, 2011: The Twentieth Century Reanalysis Project. Quart. J. Roy. Meteor. Soc., 137, 1–28.
Curry, J. A., 2008: Potential increased hurricane activity in a greenhouse warmed world. Sudden and Disruptive Climate Change, M. C. MacCracken, F. Moore, and J. C. Topping, Eds., Earthscan, 29–38.
Delworth, T. L., and M. E. Mann, 2000: Observed and simulated multidecadal variability in the Northern Hemisphere. Climate Dyn., 16, 661–676.
Dessler, A. E., and S. M. Davis, 2010: Trends in tropospheric humidity from reanalysis systems. J. Geophys. Res., 115, D19127, doi:10.1029/2010JD014192.
Emanuel, K. A., 1986: An air–sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. J. Atmos. Sci., 43, 585–604.
Emanuel, K. A., 2005: Increasing destructiveness of tropical cyclones over the past 30 years. Nature, 436, 686–688.
Emanuel, K. A., 2008: The hurricane–climate connection. Bull. Amer. Meteor. Soc., 89, ES10–ES20.
Emanuel, K. A., 2010: Tropical cyclone activity downscaled from NOAA-CIRES reanalysis, 1908–1958. J. Adv. Model. Earth Syst., 2 (1), doi:10.3894/JAMES.2010.2.1.
Emanuel, K. A., and D. S. Nolan, 2004: Tropical cyclone activity and global climate. Preprints, 26th Conf. on Hurricane and Tropical Meteorology, Miami, FL, Amer. Meteor. Soc., 240–241.
Enfield, D. B., and L. Cid-Serrano, 2010: Secular and multidecadal warmings in the North Atlantic and their relationships with major hurricane activity. Int. J. Climatol., 30, 174–184.
Enfield, D. B., A. M. Mestas-Nuñez, and P. J. Trimble, 2001: The Atlantic multidecadal oscillation and its relationship to rainfall and river flows in the continental U.S. Geophys. Res. Lett., 28, 2077–2080.
Goldenberg, S. B., C. W. Landsea, A. M. Mestas-Nuñez, and W. M. Gray, 2001: The recent increase in Atlantic hurricane activity: Causes and implications. Science, 293, 474–478.
Gray, S. T., L. J. Graumlich, J. L. Betancourt, and G. T. Pederson, 2004: A tree-ring based reconstruction of the Atlantic multidecadal oscillation since 1567 A.D. Geophys. Res. Lett., 31, L12205, doi:10.1029/2004GL019932.
Gray, W. M., 1968: Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev., 96, 669–700.
Gray, W. M., 1975: Tropical cyclone genesis. Colorado State University Department of Atmospheric Science Paper 234, 121 pp.
Hagos, A. M., and K. H. Cook, 2008: Ocean warming and late-twentieth-century Sahel drought and recovery. J. Climate, 21, 3797–3814.
Hodson, D. L. R., R. T. Sutton, C. Cassou, N. Keenlyside, Y. Okumura, and T. Zhou, 2009: Climate impacts of recent multidecadal changes in Atlantic Ocean sea surface temperature: A multimodel comparison. Climate Dyn., 34, 1041–1058.
Holland, G. J., 2007: Misuse of landfall as a proxy for Atlantic tropical cyclone activity. Eos, Trans. Amer. Geophys. Union, 88, 349–350.
Holland, G. J., and P. J. Webster, 2007: Heightened tropical cyclone activity in the North Atlantic: Natural variability or climate trend? Philos. Trans. Roy. Soc., 365, 2695–2716.
Hoyos, C. D., P. A. Agudelo, P. J. Webster, and J. A. Curry, 2006: Deconvolution of the factors contributing to the increases in global hurricane intensity. Science, 312, 94–97.
Jiang, X., M. Zhao, and D. E. Waliser, 2012: Modulation of tropical cyclones over the eastern Pacific by the intraseasonal variability simulated in an AGCM. J. Climate, 25, 6524–6538.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471.
Knaff, J. A., 1997: Implications of summertime sea level pressure anomalies in the tropical Atlantic region. J. Climate, 10, 789–804.
Knight, J. R., R. J. Allan, C. K. Folland, M. Vellinga, and M. E. Mann, 2005: A signature of persistent natural thermohaline circulation cycles in observed climate. Geophys. Res. Lett., 32, L20708, doi:10.1029/2005GL024233.
Knight, J. R., C. K. Folland, and A. A. Scaife, 2006: Climate impacts of the Atlantic multidecadal oscillation. Geophys. Res. Lett., 33, L17706, doi:10.1029/2006GL026242.
Kossin, J. P., and D. J. Vimont, 2007: A more general framework for understanding Atlantic hurricane variability and trends. Bull. Amer. Meteor. Soc., 88, 1767–1781.
Landsea, C. W., 2005: Meteorology: Hurricanes and global warming. Nature, 438, E11–E12.
Landsea, C. W., 2007: Counting Atlantic tropical cyclones back to 1900. Eos, Trans. Amer. Geophys. Union, 88, 197–202.
Landsea, C. W., G. A. Vecchi, L. Bengtsson, and T. R. Knutson, 2010: Impact of duration thresholds on Atlantic tropical cyclone counts. J. Climate, 23, 2508–2519.
Mann, M. E., and K. A. Emanuel, 2006: Atlantic hurricane trends linked to climate change. Eos, Trans. Amer. Geophys. Union, 87, 233–244.
Mann, M. E., K. A. Emanuel, G. J. Holland, and P. J. Webster, 2007a: Atlantic tropical cyclones revisited. Eos, Trans. Amer. Geophys. Union, 88, 349–350.
Mann, M. E., T. A. Sabbatelli, and U. Neu, 2007b: Evidence for a modest undercount bias in early historical Atlantic tropical cyclone counts. Geophys. Res. Lett., 34, L22707, doi:10.1029/2007GL031781.
McGauley, M. G., and D. S. Nolan, 2011: Measuring environmental favorability for tropical cyclogenesis by statistical analysis of threshold parameters. J. Climate, 24, 5968–5997.
Menkes, C. E., M. Lengaigne, P. Marchesiello, N. C. Jourdain, E. M. Vincent, J. Lefevre, F. Chauvin, and J. Royer, 2012: Comparison of tropical cyclogenesis indices on seasonal to interannual timescales. Climate Dyn., 38, 301–321.
Murakami, H., and B. Wang, 2010: Future change of North Atlantic tropical cyclone tracks: Projection by a 20-km-mesh global atmospheric model. J. Climate, 23, 2699–2721.
Nolan, D. S., E. D. Rappin, and K. A. Emanuel, 2006: Could hurricanes form from random convection in a warmer world? Extended Abstracts, 27th Conf. on Hurricanes and Tropical Meteorology, Monterey, CA, Amer. Meteor. Soc., 1C.8. [Available online at https://ams.confex.com/ams/27Hurricanes/techprogram/paper_107936.htm.]
Onogi, K., and Coauthors, 2005: JRA-25: Japanese 25-Year Re-Analysis Project—Progress and status. Quart. J. Roy. Meteor. Soc., 131, 3259–3268.
Pielke, R. A., C. Landsea, M. Mayfield, J. Laver, and R. Pasch, 2005: Hurricanes and global warming. Bull. Amer. Meteor. Soc., 86, 1571–1575.
Rienecker, M., and Coauthors, 2011: MERRA: NASA's Modern-Era Retrospective Analysis for Research and Applications. J. Climate, 24, 3624–3648.
Shapiro, L. J., and S. B. Goldenberg, 1998: Atlantic sea surface temperatures and tropical cyclone formation. J. Climate, 11, 578–590.
Simmons, A., S. Uppala, and D. Dee, 2007: Update on ERA-Interim. ECMWF Newsletter, No. 111, ECMWF, Reading, United Kingdom, 5. [Available online at http://www.ecmwf.int/publications/newsletters/pdf/111.pdf.]
Smith, T. M., R. W. Reynolds, T. C. Peterson, and J. Lawrimore, 2008: Improvements to NOAA's historical merged land–ocean surface temperature analysis (1880– 2006). J. Climate, 21, 2283–2296.
Solomon, S., D. Qin, M. Manning, M. Marquis, K. Averyt, M. M. B. Tignor, H. L. Miller Jr., and Z. Chen, Eds., 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.
Sutton, R. T., and D. L. R. Hodson, 2005: Atlantic Ocean forcing of North American and European summer climate. Science, 309, 115–118.
Ting, M., Y. Kushnir, R. Seager, and C. Li, 2009: Forced and internal twentieth-century SST trends in the North Atlantic. J. Climate, 22, 1469–1481.
Tippett, M. K., S. J. Camargo, and A. H. Sobel, 2011: A Poisson regression index for tropical cyclone genesis and the role of large-scale vorticity in genesis. J. Climate, 24, 2335–2357.
Trenberth, K. E., and D. J. Shea, 2006: Atlantic hurricanes and natural variability in 2005. Geophys. Res. Lett., 33, L12704, doi:10.1029/2006GL026894.
Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 2961–3012.
Vecchi, G. A., and B. J. Soden, 2007: Increased tropical Atlantic wind shear in model projections of global warming. Geophys. Res. Lett., 34, L08702, doi:10.1029/2006GL028905.
Vecchi, G. A., and T. R. Knutson, 2008: On estimates of historical North Atlantic tropical cyclone activity. J. Climate, 21, 3580–3600.
Villarini, G., G. A. Vecchi, T. R. Knutson, and J. A. Smith, 2011: Is the recorded increase in short-duration North Atlantic tropical storms spurious? J. Geophys. Res., 116, D10114, doi:10.1029/2010JD015493.
Wang, C., S. K. Lee, and D. B. Enfield, 2008: Atlantic warm pool acting as a link between Atlantic multidecadal oscillation and Atlantic tropical cyclone activity. Geochem. Geophys. Geosyst., 9, Q05V03, doi:10.1029/2007GC001809.
Wang, C., S. Dong, A. T. Evan, G. R. Foltz, and S. Lee, 2012: Multidecadal covariability of North Atlantic sea surface temperature, African dust, Sahel rainfall, and Atlantic hurricanes. J. Climate, 25, 5404–5415.
Webster, P. J., G. J. Holland, J. A. Curry, and H.-R. Chang, 2005: Changes in tropical cyclone number, duration, and intensity in a warming environment. Science, 309, 1844–1846.
Wu, L., and B. Wang, 2008: What has changed the proportion of intense hurricanes in the last 30 years? J. Climate, 21, 1432–1439.
Wu, L., L. Tao, and Q. Ding, 2010: Influence of sea surface warming on environmental factors affecting long-term changes of Atlantic tropical cyclone formation. J. Climate, 23, 5978–5989.
Zhang, R., and T. L. Delworth, 2006: Impact of Atlantic multidecadal oscillations on India/Sahel rainfall and Atlantic hurricanes. Geophys. Res. Lett., 33, L17712, doi:10.1029/2006GL026267.