• Bister, M., and K. A. Emanuel, 1998: Dissipative heating and hurricane intensity. Meteor. Atmos. Phys., 65, 233240.

  • Dengler, K., 1997: A numerical study of the effects of land proximity and changes in sea surface temperature on hurricane tracks. Quart. J. Roy. Meteor. Soc., 123, 13071321.

    • Search Google Scholar
    • Export Citation
  • Dutton, J. F., C. J. Poulsen, and J. L. Evans, 2000: The effect of global climate change on the regions of tropical convection in CSM1. Geophys. Res. Lett., 27, 30493052.

    • Search Google Scholar
    • Export Citation
  • Emanuel, K. A., 1986: An air–sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. J. Atmos. Sci., 43, 585604.

    • Search Google Scholar
    • Export Citation
  • Gray, W. M., 1968: Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev., 96, 669700.

  • Johnson, N. C., and S.-P. Xie, 2010: Changes in the sea surface temperature threshold for tropical convection. Nat. Geosci., 3, 842845, doi:10.1038/ngeo1008.

    • Search Google Scholar
    • Export Citation
  • Knapp, K. R., M. C. Kruk, D. H. Levinson, and E. J. Gibney, 2009: Archive compiles new resource for global tropical cyclone research. Eos, Trans. Amer. Geophys. Union, 90, doi:10.1029/2009EO060002.

    • Search Google Scholar
    • Export Citation
  • Knutson, T. R., J. J. Sirutis, S. T. Garner, G. A. Vecchi, and I. M. Held, 2008: Simulated reduction in Atlantic hurricane frequency under twenty-first-century warming conditions. Nat. Geosci., 1, 359364.

    • Search Google Scholar
    • Export Citation
  • McBride, J. L., and K. Fraedrich, 1995: CISK: A theory for the response of tropical convective complexes to variations in sea surface temperature. Quart. J. Roy. Meteor. Soc., 121, 783796.

    • Search Google Scholar
    • Export Citation
  • Miller, B. I., 1958: On the maximum intensity of hurricanes. J. Meteor., 15, 184195.

  • Palmén, E., 1948: On the formation and structure of tropical hurricanes. Geophysica, 3, 2638.

  • Reynolds, R. W., T. M. Smith, C. Liu, D. B. Chelton, K. S. Casey, and M. G. Schlax, 2007: Daily high-resolution blended analyses for sea surface temperature. J. Climate, 20, 54735496.

    • Search Google Scholar
    • Export Citation
  • Rodgers, E. B., and R. F. Adler, 1981: Tropical rainfall characteristics as determined from a satellite passive microwave radiometer. Mon. Wea. Rev., 109, 506521.

    • Search Google Scholar
    • Export Citation
  • Rodgers, E., W. Olson, J. Halverson, J. Simpson, and H. Pierce, 2000: Environmental forcing of Supertyphoon Paka’s (1997) latent heat structure. J. Appl. Meteor., 39, 19832006.

    • Search Google Scholar
    • Export Citation
  • Sitkowski, M., and G. M. Barnes, 2009: Low-level thermodynamic, kinematic, and reflectivity fields of Hurricane Guillermo (1997) during rapid intensification. Mon. Wea. Rev., 137, 645663.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., and B. J. Soden, 2007: Effect of remote sea surface temperature change on tropical cyclone potential intensity. Nature, 450, 10661070, doi:10.1038/nature06423.

    • Search Google Scholar
    • Export Citation
  • Zehr, R. M., 1992: Tropical cyclogenesis in the western North Pacific. NOAA Tech. Rep. NESDIS 61, Dept. of Commerce, Washington, DC, 181 pp.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) Percentage occurrences of SSTs during tropical cyclogenesis corresponding to each 1°C bin, and accumulated percentages (solid line), (b) percentage occurrences of all SST values along TC tracks within 35° latitude of the equator and accumulated percentages (dashed line), and (c) comparison of accumulated percentages.

  • View in gallery

    SSTs and absolute latitudes of TCs found to undergo tropical cyclogenesis while located over SST < 25.5°C (the 26°C bin). The point styles correspond to those shown in Table 1, as discussed in the text. As per the discussion, the four systems denoted by triangles are the only cases of unambiguous tropical cyclone formation below 25.5°C.

  • View in gallery

    (a) Percentage occurrences of SST48s corresponding to each 1°C bin, and accumulated percentages (solid line). (b) Accumulated percentages for the SST48s (thick solid line) compared with the SST at genesis (thin solid line) and all SST values along TC tracks within 35° latitude of the equator (dashed line).

  • View in gallery

    (a) Percentage of observations where SST < 25.5°C at the TC genesis point (solid line) and percentage of observations where SST48 < 25.5°C (dashed line), over a range of threshold MSW values that define the point of TC genesis. (b) As in (a), but with threshold SST < 26.5°C.

  • View in gallery

    (a) Accumulated percentages of SSTs at the point of TC genesis corresponding to each 1°C bin for TCs that occurred during the period September 1981 to June 1995 (solid line) and during July 1995 to December 2008 (dashed line). (b) As in (a), but with the SST48s.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 541 334 28
PDF Downloads 389 242 28

The Threshold Sea Surface Temperature Condition for Tropical Cyclogenesis

View More View Less
  • 1 Centre for Australian Weather and Climate Research, Bureau of Meteorology, Melbourne, Australia
Full access

Abstract

The analyzed value of sea surface temperature (SST) is examined for all global cases of tropical cyclone formation within 35° latitude of the equator over the period 1981–2008. It is found that 98.3% of formations occur at SST values exceeding 25.5°C. This practical threshold is relatively insensitive to the exact value of maximum wind speed used to define formation. The threshold is sensitive, however, to short-term variations in SST during development. By expanding the time period such that thresholds are calculated based on the maximum SST during the 48-h period leading to genesis, it is found that 99.5% of cyclone formations occur above 25.5°C. It is also found that tropical cyclones form over a narrow temperature range with 90.4% forming over SSTs between 27.5° and 30.5°C when the 48-h period is considered. Without consideration of the 48-h period, an SST threshold of 25.5°C is representative with less than 2% of formations occurring below this value. When the 48-h period is considered, 26.5°C is the equivalent threshold. The response to warming of the global oceans is also examined. Dividing the 27-yr dataset into an earlier versus a later 13.5-yr period, positive but small changes (+0.2°C) occur in the mean formation temperature. There is no detectable shift of the threshold temperature toward a higher value.

Corresponding author address: Dr. Richard A. Dare, Centre for Australian Weather and Climate Research, Bureau of Meteorology, GPO Box 1289, Melbourne VIC 3001, Australia. E-mail: r.dare@bom.gov.au

Abstract

The analyzed value of sea surface temperature (SST) is examined for all global cases of tropical cyclone formation within 35° latitude of the equator over the period 1981–2008. It is found that 98.3% of formations occur at SST values exceeding 25.5°C. This practical threshold is relatively insensitive to the exact value of maximum wind speed used to define formation. The threshold is sensitive, however, to short-term variations in SST during development. By expanding the time period such that thresholds are calculated based on the maximum SST during the 48-h period leading to genesis, it is found that 99.5% of cyclone formations occur above 25.5°C. It is also found that tropical cyclones form over a narrow temperature range with 90.4% forming over SSTs between 27.5° and 30.5°C when the 48-h period is considered. Without consideration of the 48-h period, an SST threshold of 25.5°C is representative with less than 2% of formations occurring below this value. When the 48-h period is considered, 26.5°C is the equivalent threshold. The response to warming of the global oceans is also examined. Dividing the 27-yr dataset into an earlier versus a later 13.5-yr period, positive but small changes (+0.2°C) occur in the mean formation temperature. There is no detectable shift of the threshold temperature toward a higher value.

Corresponding author address: Dr. Richard A. Dare, Centre for Australian Weather and Climate Research, Bureau of Meteorology, GPO Box 1289, Melbourne VIC 3001, Australia. E-mail: r.dare@bom.gov.au

1. Introduction

It has been accepted for decades that development and maintenance of tropical cyclones (TCs) require a warm ocean surface underlying the storm to act as a source of energy. Palmén (1948) identified a minimum threshold value of sea surface temperature (SST) of 26°C. Miller (1958) stated that a warm ocean is a necessary, but not sufficient, ingredient for major intensification of TCs. Emanuel (1986) linked SST to the maximum potential intensity of TCs. McBride and Fraedrich (1995) proposed the underlying SST dominated the moisture budget of their analytical (conditional instability of the second kind) CISK-type model, such that a fast-mode growth or instability occurred above a threshold SST value.

There have been countless references in the literature to threshold values of SST of around 26.5°C for the formation of TCs (e.g., Gray 1968; Dengler 1997; Rodgers et al. 2000). This paper examines the threshold concept in the era of modern datasets by examining the value of analyzed SST for all individual cases of tropical cyclone formation equatorward of 35° latitude over the period 1981–2008.

It is recognized here that SST is just one of several factors influencing formation and development of TCs. Thus, rather than an absolute value of SST, it may be the value relative to other environmental parameters that forms the necessary condition for cyclone formation. For example, maximum potential intensity theory (Emanuel 1986; Bister and Emanuel 1998) would imply that thermodynamic efficiency may be the controlling thermodynamic parameter. Vecchi and Soden (2007) suggest that the difference between local SST and zonal-mean or all-tropics-mean SST may be the controlling parameter. Despite this qualification, the validity of the concept of a SST threshold and its possible value is worthy of examination. In the following section we outline the data and methods, followed in section 3 by results for cyclone formation and cyclone occurrence as a function of SST. Section 4 examines the events that occur below the 25.5°C threshold and as a result expands the analysis to include the highest SST encountered in the 48 h prior to formation. Sensitivity to the wind speed threshold used to define TC formation is examined in section 5. Section 6 inspects the data for temporal trends in the threshold SST value. Conclusions are presented in section 7.

2. Data and methods

The International Best Track (IBTrACS) dataset (Knapp et al. 2009) is used to define 6-hourly observations of TCs in all 6 TC basins over the globe. As this collection also contains storms at higher (nontropical) latitudes and storms with relatively weak circulations, the data are filtered so that the following data are removed: 1) observations poleward of 35° latitude (to exclude storms with baroclinic features), 2) storms that do not contain at least one observation where the 10-min-averaged maximum sustained wind (MSW) reaches or exceeds 17 m s−1, 3) observations identified in the IBTrACS as extratropical or subtropical. These three rules discount 750 cyclones, or approximately 25%, leaving 2217 TCs for our analysis.

SSTs corresponding to each 6-hourly TC observation are obtained by interpolating from the National Oceanic and Atmospheric Administration/National Climatic Data Center (NOAA/NCDC) SST dataset (Reynolds et al. 2007). This dataset has a spatial resolution of 0.25° × 0.25° and a temporal resolution of 1 day. One of the motivations for this study is to utilize the high quality SST and TC data collected during the recent decades of the satellite era. The data used here are from September 1981 to December 2008. Separate tropical cyclone basins were examined, but no major interbasin differences were found. Hence, we present here results for the entire globe.

3. Results

SST values corresponding to the TC observations were stratified into bins with intervals of 1°C. For example, the 26°C bin contains SST values defined by the following range: 25.5°C ≤ SST < 26.5°C. TC genesis is defined here as the first point along a TC’s track where MSW ≥ 17 m s−1. The percentage occurrences of SSTs corresponding to TC genesis (Fig. 1a) show that the majority of SSTs are greater than the 26°C bin. Approximately 5% of the distribution corresponds to the 26°C bin. The cumulative line shows that approximately 1.7% of the genesis points (37 TCs) occur over SSTs less than 25.5°C (i.e., below the 26°C bin). In comparison, Fig. 1b represents the percentage occurrence of SSTs corresponding to all 6-h observations from all TCs considered in this study. As would be expected, given that TCs generally translate poleward where SSTs are generally lower, this distribution of SSTs shows that approximately 12.5% of observations occur over SSTs less than 25.5°C. This percentage would be larger (18.2%) if observations poleward of latitude 35° and systems of subtropical origin had also been considered in this study. The contrast between SST at genesis and SST corresponding to all points is illustrated more clearly in Fig. 1c, with cumulative percentages displayed side by side.

Fig. 1.
Fig. 1.

(a) Percentage occurrences of SSTs during tropical cyclogenesis corresponding to each 1°C bin, and accumulated percentages (solid line), (b) percentage occurrences of all SST values along TC tracks within 35° latitude of the equator and accumulated percentages (dashed line), and (c) comparison of accumulated percentages.

Citation: Journal of Climate 24, 17; 10.1175/JCLI-D-10-05006.1

4. Analysis and discussion

In this section the 37 genesis cases identified as occurring over SSTs less than 25.5°C (the lower limit of the 26°C bin) are examined. The details for these cases are given in Table 1, with corresponding points plotted as a function of latitude and SST in Fig. 2. The far-right column of the table and the symbols in Fig. 2 correspond to the status of the formation process.

Table 1.

Details of 37 TCs that were detected objectively to undergo tropical cyclogenesis while over SST < 25.5°C (i.e., below the 26°C bin). The symbols shown in the far-right-hand side column correspond to those plotted in Fig. 2. Basin acronyms are North Atlantic (NA), east North Pacific (ENP), South Pacific (SP), south Indian Ocean (SI), and west North Pacific (WNP). M denotes “Missing,” indicating that no information was recorded about the storm during the 48 hours prior to the genesis point.

Table 1.
Fig. 2.
Fig. 2.

SSTs and absolute latitudes of TCs found to undergo tropical cyclogenesis while located over SST < 25.5°C (the 26°C bin). The point styles correspond to those shown in Table 1, as discussed in the text. As per the discussion, the four systems denoted by triangles are the only cases of unambiguous tropical cyclone formation below 25.5°C.

Citation: Journal of Climate 24, 17; 10.1175/JCLI-D-10-05006.1

Based on inspection, we discount the following six systems from being true tropical formations. For system number 8 (asterisk in Fig. 2) there are multiple and overlapping observations of this TC in the IBTrACS dataset. When the two sets of observations for this TC are combined, the SST at genesis is found to be 28°C. Cyclone numbers 4, 9, and 18 (+ symbol) all occurred at relatively poleward latitudes and contained only five, five, and eight 6-hourly observations each, respectively. They also had no prior recorded history and moved rapidly to the 35° latitude threshold applied here. TC number 28 also had only 3 observations, meaning it existed for less than a day (marked by “○”). The TC number 29 (marked by “×”) acquired an MSW of 17 m s−1 over an SST of 19°C. However, it has a long history covering several days during which it achieved a MSW of 16 m s−1 while over an SST of 26.2°C. Given that it is difficult to accurately measure MSWs and that the MSWs in this study are 10-min averages, and therefore generally less than the 1-min averages used by some TC centers around the globe, it is sensible to conclude this cyclone did previously undergo genesis over SST ≥ 25.5°C.

Of the remaining 31 cases, all reach MSW > 17 m s−1 over SST < 25.5°C. This represents 1.4% of the 2217 TCs considered in this study. The second-to-right column of Table 1 shows the maximum SST over the preceding 48 h along the preformation track, or at formation location when there is no earlier recorded track (hereafter, referred to as SST48). This reveals that 27 of the 31 cases had existed over SSTs ≥ 25.5°C at some time during the 48 h leading up to genesis. The reduction in SST experienced by the TC may be due to translation of the storm over cooler waters or cooling of the ocean surface induced by the circulation. Either way, these data show that the storms intensified while experiencing a reduction in SST. Following the supply of energy, we speculate that there was a delay in the intensification of the storm, a possibility discussed previously by Miller (1958), Rodgers and Adler (1981), and Sitkowski and Barnes (2009). This is also consistent with Zehr’s (1992) concept of genesis occurring over several days.

The presence of a higher SST in the 48 h prior to genesis is not limited to genesis cases occurring over SST < 25.5°C. The distribution of percentage occurrence of SST48s for all TCs (Fig. 3a) shows a normal-type distribution centered on 29°C, with less of a tail toward lower SST values than the two distributions examined previously in Fig. 1. In fact, 90% of the formations occur in the 3 bins centered at 28°, 29°, and 30°C. A comparison between accumulated percentage occurrences for SST at genesis, SST corresponding to all points, and SST48 is shown in Fig. 3b. The latter contains the highest SST values, as would be expected, with just 0.4% of the data occurring for SST < 25.5°C (cf. 1.7% and 12.5% for SST at genesis and all SST data, respectively). While 6.7% of the observations occur below 26.5°C at the point of genesis, just 1.4% of the observations lie below 26.5°C when considering the SST48. This means that 26.5°C is an acceptable threshold when the 48-h period is considered, but 25.5°C is more acceptable as a threshold when the single genesis point is used.

Fig. 3.
Fig. 3.

(a) Percentage occurrences of SST48s corresponding to each 1°C bin, and accumulated percentages (solid line). (b) Accumulated percentages for the SST48s (thick solid line) compared with the SST at genesis (thin solid line) and all SST values along TC tracks within 35° latitude of the equator (dashed line).

Citation: Journal of Climate 24, 17; 10.1175/JCLI-D-10-05006.1

Accepting the delay in intensification as a valid explanation leaves just 4 cases in which TC genesis occurred over SST < 25.5°C, marked by triangles in the table and Fig. 2. Three of these (Nicole, Epsilon, and Zeta) occurred very late in the North Atlantic season (24 November, 29 November, and 30 December, respectively), which explains the lower-than-usual SSTs. Arlene underwent genesis during June 1999, also in the North Atlantic basin, and it may be the best example in our dataset of TC genesis over SST < 25.5°C. It is interesting that these four cases all occurred in the North Atlantic basin, where observations are of a high quality.

5. Sensitivity to MSW threshold

In the above analysis, a 10-min-averaged MSW threshold of 17 m s−1 was used to define TC formation. However, as different thresholds are used in different basins, a range of MSW threshold values is considered in this section. For example, in the North Atlantic basin, a value equivalent to approximately 15 m s−1 is used to define a “tropical storm.” The percentages of TC formations that occur over SSTs < 25.5°C are found to generally increase as MSW increases (Fig. 4a, solid line). When considering SST48, the percentage of SSTs < 25.5°C is fairly consistent at around 0.5% (Fig. 4a, dashed line). These SST48s also correspond to consistently smaller percentages than those for SSTs at the point of formation (solid line) over the entire range of MSWs considered here.

Fig. 4.
Fig. 4.

(a) Percentage of observations where SST < 25.5°C at the TC genesis point (solid line) and percentage of observations where SST48 < 25.5°C (dashed line), over a range of threshold MSW values that define the point of TC genesis. (b) As in (a), but with threshold SST < 26.5°C.

Citation: Journal of Climate 24, 17; 10.1175/JCLI-D-10-05006.1

The sensitivity of the percentages of SSTs below the 26.5°C threshold are also examined (Fig. 4b). In agreement with Fig. 4a, the percentages vary with the choice of MSW. Furthermore, the percentages are lower in Fig. 4a than in Fig. 4b over the range of MSW thresholds. Only for a MSW threshold greater than approximately 16 m s−1 does the percentage corresponding to the SST48 < 26.5°C (Fig. 4b, dashed line) fall below that corresponding to the SST < 25.5°C at the genesis point (Fig. 4a, solid line).

Overall, percentage changes corresponding to the relevant range of MSW thresholds (15–17 m s−1) are small for both the 25.5° and 26.5°C thresholds (0.6% and 1.6%, respectively) for the SST at the point of formation. Changes to the percentages associated with SST48 data over the same range of MSW thresholds are even smaller. In conclusion, the choice of MSW to define the storm has little impact on the results.

6. Climate change

It has been suggested that the threshold values of SST may be a function of current climate and that the actual threshold may shift toward a higher value in a warmer climate (Dutton et al. 2000; Knutson et al. 2008; Johnson and Xie 2010). Here, distributions of SST are compared for the two halves of the dataset, noting however, that the 27-yr period of the dataset is probably not adequate to provide a reliable indication of change in the climatology. Accumulated percentages of SST occurrence corresponding to TC genesis points for the first half (September 1981–June 1995) and second half (July 1995–December 2008) of the period are shown by solid and dashed lines, respectively, in Fig. 5a. The corresponding accumulated percentages of the SST48s are shown in Fig. 5b.

Fig. 5.
Fig. 5.

(a) Accumulated percentages of SSTs at the point of TC genesis corresponding to each 1°C bin for TCs that occurred during the period September 1981 to June 1995 (solid line) and during July 1995 to December 2008 (dashed line). (b) As in (a), but with the SST48s.

Citation: Journal of Climate 24, 17; 10.1175/JCLI-D-10-05006.1

The number of formations occurring above 25.5°C decreased from 99.0% in the earlier period to 97.7% in the later period. The reason may simply be due to changes in operational procedures in the presence of higher-quality real-time satellite data. Since the change in threshold percentages is opposite to that expected in a warmer climate, it is concluded that there is no increasing threshold signal in this dataset. However, there is a shift toward warmer temperatures for the majority of formation events. For example, the percentage of formations occurring at temperatures exceeding 27.5°C increased from 36.7% in the earlier period to 48.1% in the later period. The mean temperature of formation increased from 28.1° to 28.3°C. In summary, there is a detectable shift toward formation at higher values of SST in the later period, but the effect is small and of the order of 0.2°C change in mean formation temperature. There is no detected shift of the formation threshold toward a higher value.

7. Conclusions

Analysis with current data (1981–2008) confirms the existence of a sea surface temperature threshold for tropical cyclone formation. New findings are as follows:

  1. Globally, more than 93% of cyclone formations occurred at SST values exceeding 26.5°C and more than 98% occurred at SST values exceeding 25.5°C. Although a small number of TCs do form over lower SSTs, the choice for a practical threshold lies between approximately 25.5° and 26.5°C, as suggested by Fig. 3b and consistent with past work.

  2. In contrast, there is no strong threshold for tropical cyclone existence, with more than 77% of cyclone locations occurring at SSTs exceeding 26.5°C, and more than 87% at SSTs exceeding 25.5°C.

  3. When the SST window is expanded to include the maximum SST in the 48 h prior to formation, the threshold is even stronger with 99.5% of global formations occurring with a 48-h SST value greater than 25.5°C, and more than 98% greater than 26.5°C.

  4. Considering tropical cyclone formation as a process that takes 48 h to occur, TC formation is found to occur over a narrow span of 48-h-maximum SSTs with more than 90% in the 3°C range (27.5°–30.5°C).

  5. The percentage of TC formations occurring above some threshold SST value is not strongly sensitive to the MSW threshold used to define TC formation.

  6. Dividing the data into two consecutive 13.5-yr datasets, there is a shift toward TC formation occurring at higher SST values, with the mean genesis temperature increasing by 0.2°C. Given the small period of data (27 yr) and the changes in data quality and operational procedures that have occurred in recent decades, no attempt has been made to give a statistical significance to this result.

  7. There is no detectable shift of the SST formation threshold value of 25.5°C toward a warmer value. In fact there were more TC formations at temperatures below this threshold in the later half of the time series.

Acknowledgments

We thank Dr. Kevin Tory for reading the manuscript and providing constructive feedback. We are also grateful to the two anonymous reviewers for their positive and constructive criticisms. This work is supported by the Indian Ocean Climate Initiative funded by the Government of Western Australia.

REFERENCES

  • Bister, M., and K. A. Emanuel, 1998: Dissipative heating and hurricane intensity. Meteor. Atmos. Phys., 65, 233240.

  • Dengler, K., 1997: A numerical study of the effects of land proximity and changes in sea surface temperature on hurricane tracks. Quart. J. Roy. Meteor. Soc., 123, 13071321.

    • Search Google Scholar
    • Export Citation
  • Dutton, J. F., C. J. Poulsen, and J. L. Evans, 2000: The effect of global climate change on the regions of tropical convection in CSM1. Geophys. Res. Lett., 27, 30493052.

    • Search Google Scholar
    • Export Citation
  • Emanuel, K. A., 1986: An air–sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. J. Atmos. Sci., 43, 585604.

    • Search Google Scholar
    • Export Citation
  • Gray, W. M., 1968: Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev., 96, 669700.

  • Johnson, N. C., and S.-P. Xie, 2010: Changes in the sea surface temperature threshold for tropical convection. Nat. Geosci., 3, 842845, doi:10.1038/ngeo1008.

    • Search Google Scholar
    • Export Citation
  • Knapp, K. R., M. C. Kruk, D. H. Levinson, and E. J. Gibney, 2009: Archive compiles new resource for global tropical cyclone research. Eos, Trans. Amer. Geophys. Union, 90, doi:10.1029/2009EO060002.

    • Search Google Scholar
    • Export Citation
  • Knutson, T. R., J. J. Sirutis, S. T. Garner, G. A. Vecchi, and I. M. Held, 2008: Simulated reduction in Atlantic hurricane frequency under twenty-first-century warming conditions. Nat. Geosci., 1, 359364.

    • Search Google Scholar
    • Export Citation
  • McBride, J. L., and K. Fraedrich, 1995: CISK: A theory for the response of tropical convective complexes to variations in sea surface temperature. Quart. J. Roy. Meteor. Soc., 121, 783796.

    • Search Google Scholar
    • Export Citation
  • Miller, B. I., 1958: On the maximum intensity of hurricanes. J. Meteor., 15, 184195.

  • Palmén, E., 1948: On the formation and structure of tropical hurricanes. Geophysica, 3, 2638.

  • Reynolds, R. W., T. M. Smith, C. Liu, D. B. Chelton, K. S. Casey, and M. G. Schlax, 2007: Daily high-resolution blended analyses for sea surface temperature. J. Climate, 20, 54735496.

    • Search Google Scholar
    • Export Citation
  • Rodgers, E. B., and R. F. Adler, 1981: Tropical rainfall characteristics as determined from a satellite passive microwave radiometer. Mon. Wea. Rev., 109, 506521.

    • Search Google Scholar
    • Export Citation
  • Rodgers, E., W. Olson, J. Halverson, J. Simpson, and H. Pierce, 2000: Environmental forcing of Supertyphoon Paka’s (1997) latent heat structure. J. Appl. Meteor., 39, 19832006.

    • Search Google Scholar
    • Export Citation
  • Sitkowski, M., and G. M. Barnes, 2009: Low-level thermodynamic, kinematic, and reflectivity fields of Hurricane Guillermo (1997) during rapid intensification. Mon. Wea. Rev., 137, 645663.

    • Search Google Scholar
    • Export Citation
  • Vecchi, G. A., and B. J. Soden, 2007: Effect of remote sea surface temperature change on tropical cyclone potential intensity. Nature, 450, 10661070, doi:10.1038/nature06423.

    • Search Google Scholar
    • Export Citation
  • Zehr, R. M., 1992: Tropical cyclogenesis in the western North Pacific. NOAA Tech. Rep. NESDIS 61, Dept. of Commerce, Washington, DC, 181 pp.

    • Search Google Scholar
    • Export Citation
Save