1. Introduction

The effect of the ocean on tropical cyclone (TC) genesis and maintenance has been well known for decades. The ocean provides the necessary energy to establish and maintain deep convection (Byers 1944; Palmen 1948; Riehl 1954; Miller 1958; Malkus and Riehl 1960). Recent studies conducted by Shay et al. (2000) and Bosart et al. (2000) have also shown that in some instances, warm upper-ocean features can significantly impact TC intensity. While findings from these case studies are significant, it is still unclear how (and to what extent) variations in upper-ocean thermal structure directly impact changes in storm intensity. Tropical cyclone intensity change is a complex and interactive nonlinear process that often involves several competing or synergistic factors (Riehl 1948; 1950; Miller 1958; Sadler 1978; Gray 1979; Holland and Merrill 1984; Emanuel 1986, 1988; DeMaria and Pickle 1988; Molinari and Vollaro 1990; Willoughby and Black 1996; Holland 1997; DeMaria and Kaplan 1994; 1999; Kaplan and DeMaria 1999; Bosart et al. 2000; Shay et al. 2000). Identifying the quantitative impact a physical process has on intensity change is an arduous task and one that can only be attempted using controlled numerical methodology. Ongoing coupled ocean–atmosphere TC modeling efforts are works in progress; many of the numerical routines and parameterizations (e.g., data initialization, grid resolution, turbulent fluxes, atmospheric microphysics, etc.) used within the rarely observed high-wind storm environment still require significant improvement.

An additional stumbling block confronting TC modelers is data verification of the upper ocean and atmospheric boundary layer (ABL) hurricane environments. Prior to 1997, accurate depictions of inner-core ABL thermodynamic and kinematic structure were essentially unknown. Also, since many of the earlier TC ocean response studies concentrated on studying the poststorm, cold wake region of the hurricane (Federov et al. 1977; Pudov et al. 1978; Pudov 1980; Price 1981), accurate depictions of the TC upper-ocean eyewall region have been exceedingly rare over the past 30 years (Black 1983; Black et al. 1988; Shay et al. 1992; Black and Holland 1995). As a result, numerical attempts to initialize and verify this rarely observed ocean environment have relied on prestorm ambient sea surface temperatures (SST) ahead of the system and post-storm cold wake SSTs, typically valid several days after TC passage (Bender et al. 1993; Bender et al. 2000). However, recent experiments conducted during the Hurricane Research Division's (HRD) annual field program have helped fill in these atmospheric and oceanic data void regions by using Global Positioning System (GPS) dropsondes and airborne expendable bathythermographs (AXBT).

To accurately document TC-induced SST change, this study will use upper-ocean data obtained from AXBTs during 1997–2002 HRD field experiments along with fixed and drifting buoy observations over the 1975–2002 period. It is believed that these (multistorm) observations and analyses will improve the representation of SST cooling patterns typically observed in the high-wind hurricane environment. These rare observations will serve as “ocean truth” for coupled modeling efforts attempting to simulate TC-induced SST change directly under the storm.

2. Research goals

Earlier observational studies have documented the significant impact hurricanes can have on the vertical and horizontal structure of the upper-ocean environment (Federov et al. 1977; Pudov et al. 1978; Pudov 1980; Price 1981; Black 1983; Shay et al. 1992). Figure 1 illustrates this point and depicts ocean surface temperatures 4°–5°C cooler in the post-Georges (1998) cold wake region (relative to the surrounding, ambient ocean environment). However, it should be recognized that this ocean response analysis was constructed seven days after the storm's passage and does not necessarily represent typical SST cooling patterns observed directly under the hurricane. This is an important point since much of the ocean-to-atmosphere exchange of energy in hurricanes occurs within a relatively limited area near the eyewall. As a result, obtaining accurate representations of near-surface ocean temperatures within this critical region is of paramount importance. By using buoy observations with recent hurricane AXBT data, this study will quantify the relative magnitude and inherent variability associated with differences between ambient SSTs well ahead of the storm and SSTs observed near storm center. The authors will investigate any potential relationships between these “SST differences” and observed changes in storm intensity. In addition, the potential impact SST variability has on air–sea fluxes within the high-wind hurricane environment will also be investigated. Finally, using observations and recent findings from Cione et al. (2000), estimates for hurricane heat potential, energy extracted by the storm, and energy utilization will be made.

3. Methodology and data used

Early ocean-response studies concentrated on analyzing the horizontal and vertical thermal structure of the upper-ocean within the TC-modified, cold wake environment. In addition to these poststorm wake studies, Black (1983) conducted analyses of the upper-ocean thermal structure for a number of Atlantic, Gulf of Mexico, and eastern Pacific hurricanes between 1971 and 1980. Due to difficult observing conditions and limited opportunities, there have only been a few case studies that have attempted to document the hurricane high-wind upper-ocean environment (Black et al. 1988; Shay et al. 1992; Black and Holland 1995). As such, quantifying the interstorm variability associated with upper-ocean thermal conditions near/within the hurricane eyewall has been difficult. A primary goal of this study is to improve SST cooling estimates near the hurricane eyewall using upper-ocean measurements from many (23) hurricanes. To accomplish this, observations from the Tropical Cyclone Buoy Database (TCBD) (Cione et al. 2000) are used. The TCBD includes information on hurricane position and intensity as well as near-surface meteorological and oceanographic data from fixed and drifting platforms. The TCBD incorporates observations from over 40 hurricanes between 1975 and 2002. Most of the TCBD observations were acquired from the National Data Buoy Center quality controlled buoy archive (Gilhousen 1988, 1998).¹ In addition to the TCBD data, upper-ocean AXBT observations from nine HRD field experiments were also used. From both buoy and AXBT data, 33 along-track SST transects were obtained for 23 hurricanes (Table 1).

To document TC-induced SST change, clear definitions for both ambient and inner-core SSTs must be established. For this study, ambient SSTs (SST_A) are located well ahead of the storm center (>2° latitude) in either the right front or left front quadrant (defined in a storm-relative coordinate system). The location of SST_A is defined as the point where SST initially decreases. Inner-core SST (SST_IC) is defined as the minimum SST within a 60-km radius of the analyzed TC center. Since it is a primary goal of this research to investigate the magnitude and variability of SST change for mature tropical systems, only observations from hurricanes are included in this analysis. All storms are prelandfall, located south of 36°N, and attain hurricane intensity (i.e., maximum surface wind speed > 32 m s⁻¹) at some point during the period of observation. Whenever possible, estimates for inner-core wake SST (SST_ICW) are also included. SST_ICW is defined as the minimum SST observed in either the right rear or left rear quadrant of the storm. However, unlike typical TC-induced cold wake SST fields (Fig. 1), SST_ICW observations are located <200 km from the storm center. As such, SST_ICW observations are often located in areas of moderate surface wind (typically > 15 m s⁻¹). Due to various factors such as instrument failure, systems making landfall, and experimental design limitations, 26 of the 33 horizontal SST transects listed in Table 1 include estimates for ΔSST_ICW.

4. Horizontal variability of SST in hurricanes

The 33 (26) ΔSST_IC (ΔSST_ICW) estimates shown in Table 2 represent the difference between SST_IC(SST_ICW) and SST_A ahead of the storm. Table 2 is stratified by ΔSST_IC with associated ΔSST_ICW values listed whenever available. Figure 2 illustrates the geographic location and magnitude for all 33 ΔSST_IC estimates listed in Table 2.

The ΔSST_IC and ΔSST_ICW values depicted in Table 2 are much less than the 4°–5°C TC-induced cold wake SST reductions often observed 1–2 days after a hurricane passage (Price 1981; Mayer et al. 1981; Black 1983). In comparison, average ΔSST_IC and ΔSST_ICW are −0.7°C and −1.0°C, respectively. These values represent ∼15%–25% of the 4°–5°C cold wake reductions shown in Fig. 1 and are in reasonable agreement with early mixed layer TC–ocean modeling studies (Elsberry et al. 1976; Chang and Anthes 1978) as well as inner-core observations from Black (1983). Cumulative distributions for both ΔSST_IC and ΔSST_ICW are given in Fig. 3. These results show that both SST_IC and SST_ICW are significantly warmer than poststorm, TC cold wake SSTs. Figure 4 captures the along-track variability of SST change as a function of radial distance (RD) from the storm center for all 22 (19) ΔSST_IC (ΔSST_ICW) buoy transects listed in Table 2 (AXBT-derived ΔSST_IC and ΔSST_ICW observations are not included in Fig. 4 due to horizontal and temporal sampling limitations associated with these data). SST reductions of 1.5°C or less were noted for 21 of 22 (15 of 19) SST_IC(SST_ICW) buoy transects.

The summary shown in Table 3 includes both buoy and AXBT-derived data. Stratifying results by ΔSST_IC, upper and lower 50th percentile statistical summaries are illustrated for ΔSST_IC, ΔSST_ICW, SST_A, SST_IC, SST_ICW, the radial distance at which SST_A initially decreased (RD_A), the radial distance where SST_IC was observed (RD_IC), the radial distance where SST_ICW was measured (RD_ICW), and storm-specific parameters such as storm latitude (TC_LAT), storm intensity (TCI_WIND; TCI_BAR), and storm speed (TC_SPEED). Tests for statistical significance between upper and lower percentile means were conducted. Table 3 depicts differences between upper and lower 50th percentile means for SST_IC, SST_ICW, TC_LAT, and RD_A (bolded values). It should be noted that statistically significant findings were not found for SST_A. In fact, ambient SST showed little variability with respect to inner-core SST change. Results from Table 3 show that lower (higher) latitude events, on average, exhibited less (more) inner-core SST cooling. This result may in part be explained by climatology. It stands to reason that, on average, lower latitude events would encounter deeper, warm water upper-ocean environments. These events (assuming all other factors to be equal) should experience less upper-ocean cooling. In addition, for a given wind speed, the onset of SST cooling for the lower 50% sample would tend to occur closer to the storm center due to the presence of (relatively) deeper warm water. This may partly explain why average RD_A (i.e., the point at which SST_A first decreases) was ∼120 km closer to the storm center for the lower 50th percentile group of observations. While differences between upper and lower sample means were not found for storm speed or storm initial intensity, results from Table 3 suggest that faster moving, initially weaker storms may be more likely to experience reduced ΔSST values. These trends, while statistically inconclusive, are in agreement with earlier results (Black 1983; Bender et al. 2000; Chan et al. 2001) and are physically consistent given the reduced level of upper-ocean turbulent mixing one would expect from quicker-moving, relatively weaker tropical systems.

5. Hurricane heat potential and energy extracted by the storm

Leiper and Volgenau (1972) first defined the term “hurricane heat potential” as the integrated vertical temperature from the sea surface to the depth of the 26°C isotherm. Hurricane heat potential (also known as “upper-ocean heat content”) is denoted by Q_H and is defined as

View Expanded

where c_p is the specific heat of water at constant pressure (4178 J kg⁻¹ K⁻¹), ρ is the average density of the upper ocean (1026 kg m⁻³), and ΔT is the difference between T(z) and 26°C over the depth interval dz. The units for Q_H are given in kJ cm⁻² (=10⁷ J m⁻²), as is common in the literature.

Since much of the ocean-to-atmosphere exchange of energy in tropical cyclones occurs within the high-wind inner core, analyses and estimates presented here will focus on conditions potentially present within this important region. Exactly how much of the available Q_H is extracted by the storm within the inner core is very difficult to quantify without highly accurate, direct, and continuous measurement. However, given the storm speed, storm initial intensity, and upper-ocean thermal profile ahead of the storm, it is possible to estimate the amount of upper-ocean heat content extracted by the storm. In order to maintain consistency with earlier definitions, the inner-core upper-ocean environment is defined to be 0–60 km from the storm center. For these calculations, the initial upper-ocean temperature profile used to calculate Q_H is located 60 km ahead of the storm along the storm track. Due to the relatively short analysis period (3–13 h), coupled with the fact that the initial profile 60 km ahead of the storm is already well mixed, inner-core Q_H estimates are assumed to remain constant for this analysis. By utilizing various storm speeds and intensities, estimates for upper-ocean heat content extracted by the storm (Q_{H_ext}) and estimates for “upper-ocean energy utilization” (Q_{H_util}) can be constructed.

First, the inner-core enthalpy (sensible plus latent heat) flux (H_core) is computed using the standard bulk aerodynamic formulas:

View Expanded

where ρ is the density of air, T_A is the air temperature at 10 m, c_p is the specific heat of air at constant pressure, and L_υ is the latent heat of vaporization at a given T_A. Here, U represents the 1-min wind speed at 10 m, while q_SST and q_A are the saturation mixing ratio at the SST and the actual mixing ratio of the air at 10 m, respectively; C_h and C_e are the dimensionless coefficients of heat exchange and moisture exchange at 10 m; H_core has units of W m⁻² (=10⁻⁷ kJ cm⁻² s⁻¹). Then Q_{H_ext} and Q_{H_util} are determined from H_core by

View Expanded

where TC_transit-time is the time in seconds for a storm to travel the inner-core diameter (120 km). To first order, Q_{H_util} estimates how much of the upper-ocean heat content available to the system is actually extracted (i.e., “utilized”) by the storm. Similar to Q_H, estimates for Q_{H_ext} are given in kJ cm⁻². For this particular analysis, H_core serves as a proxy for storm intensity since it has a strong wind speed dependence.

Recent findings from Cione et al. (2000) helped better define near-surface atmospheric thermodynamic conditions typically observed within the hurricane inner-core environment. Using these estimates and assuming a fairly typical tropical Atlantic summer season Q_H value of 75 kJ cm⁻² (=7.5 × 10⁸ J m⁻²), estimates for Q_{H_util} were constructed.² A primary objective of this study is to establish a reasonable “parameter space” for Q_{H_util} within the well-mixed hurricane inner core over a wide range of possible storm speeds and intensities. Upper-ocean energy utilization as a function of storm speed and total surface enthalpy flux is illustrated in Fig. 5. Figure 5 includes a wide array of potential storm speeds (2.5–10 m s⁻¹) and inner-core surface flux values/intensities (650–2600 W m⁻²). Applying bulk aerodynamic formulas [Eq. (2)] with exchange coefficients as determined by Garratt (1977), composite analyses from Cione et al. (2000) showed that the average inner-core total surface enthalpy flux was approximately 1300 W m⁻² for a category 1 hurricane (i.e., maximum surface wind speed between 33 and 43 m s⁻¹). The full range of energy utilization for this analysis is estimated to be between 1.0% (for a tropical storm moving at 10 m s⁻¹) and 16.6% (major hurricane moving at 2.5 m s⁻¹). It should be noted that these energy utilization estimates assume that the upper-ocean profile never enters the storm's eye. If the profile were to temporarily experience reduced surface winds within the eye, the energy utilization estimates presented in Fig. 5 (and estimated above) would be reduced.

Even though these findings do not take into account all the physical processes and/or situations that could potentially come into play (such as a stationary system or warm/cold water advection near a highly baroclinic ocean front), the results nevertheless illustrate the vast energy resources available to most tropical cyclones under most storm conditions. These results suggest that for the large majority of propagating systems, the magnitude of upper-ocean heat content (Q_H) should not be a limiting factor affecting storm maintenance and/or intensification.

6. The impact of SST change on inner-core surface enthalpy flux

Results previously illustrated in Table 3 depict an average difference in inner-core SST of 0.7°C between the upper and lower 50th percentile samples. This relatively small difference in SST can potentially have a significant impact on the resulting surface enthalpy flux to the storm within the high-wind inner core. Changes to q_A, U, and/or T_A (or modifications to the exchange coefficient expressions) will also significantly impact the magnitude of H_core. However, a primary objective of this research is to isolate the impact that SST-dependent variables (SST and q_SST) potentially have on the storm's ability to extract energy from the inner-core upper-ocean environment.

Figure 6 illustrates the percent change in upper-ocean energy extracted by the storm (ΔQ_{H_ext}) as a function of inner-core SST change [where non-SST-dependent variables in Eq. (2) remain constant]. The initial values for SST_IC, storm speed (TC_speed), surface air temperature (T_A), relative humidity (RH), surface wind speed (U), and minimum sea level pressure (P) are shown at the top of the illustration and represent bulk mean values obtained from Cione et al. (2000). Changes in inner-core total surface heat flux (relative to the initial 1300 W m⁻²) are also illustrated within the body of the figure.

Figure 6 shows that, all other factors being equal, relatively small variations in inner-core SST can dramatically impact inner-core surface heat flux and, as such, the magnitude of upper-ocean energy extracted by the storm (Q_{H_ext}). Figure 6 illustrates that a +0.7°C difference in inner-core SST (i.e., average ΔSST_IC between upper and lower 50th percentile samples shown in Table 3) results in a 30% increase in the amount of upper-ocean energy extracted by the storm (Q_{H_ext}), for a category 1 hurricane. This represents a 26% increase in H_L(270 W m⁻²) and a 33% change in H_S(80 W m⁻²). On the other hand, it should be noted that total enthalpy flux changes of this magnitude (i.e., 350 W m⁻²) would not impact the total amount of upper-ocean heat content (Q_H) available to a propagating system by more than 1%–2% (Fig. 5).

The large values of total surface enthalpy flux that result from relatively modest changes in inner-core SST, coupled with the realization that most hurricanes utilize less than 10% of the upper-ocean energy available to them, accentuates the need for a shift in focus from analyzing upper-ocean heat content to accurately observing (and predicting) the short-term variability of hurricane inner-core SST conditions.

7. Linkages between SST change and TC intensity change

Earlier studies have had difficulty linking environmental SST ahead of the storm (SST_A) with subsequent changes in TC intensity. Much of this inability can be linked to the fact that SST measurements, especially satellite-derived skin temperatures, do not adequately depict thermal conditions below the surface (Reynolds 1988; Reynolds and Smith 1994). Results in Fig. 7a also depict this trend, illustrating little or no relationship between SST_A and subsequent TC intensity change (defined in all cases as the 24-h change in maximum surface wind speed centered at the time SST_IC was recorded). Figure 7b also depicts little or no relationship between hurricane heat potential (Q_H) ahead of the storm and intensity change. In contrast, the regression results shown in Figs. 7c and 7d illustrate clear relationships between SST change (ΔSST_IC and ΔSST_ICW) and subsequent changes in storm intensity. The 4.4% and 4.2% explained variances (i.e., 100r²) shown in Figs. 7a and 7b, increase to 33.4% and 42.1% in Figs. 7c and 7d, respectively, when ΔSST_IC and ΔSST_ICW are used. This marked increase in explained variance suggests that normalized differences between SST within and ahead of the storm may, under certain circumstances, be closely tied to observed changes in hurricane intensity. The relationship between reduced inner-core SST cooling and subsequent intensification is plausible since relatively small changes in inner-core SST can significantly alter surface energy fluxes within the high-wind hurricane environment (Fig. 6). Storms experiencing reduced inner-core SST cooling would have larger surface fluxes and, as a result, would be more likely to experience enhanced intensification (assuming all other factors potentially impacting hurricane intensity change to be equal).

This hypothesis suggesting a quantifiable relationship between SST change and subsequent changes in TC intensity is tested. Table 4a is a statistical summary of results found when observations were sorted by TC intensity change (ΔTCI). Similar to the definition used in Figs. 7a–d, TC intensity change is defined as the best-track-derived (Neumann et al. 1993) 24-h change in maximum surface wind speed centered at the time SST_IC was obtained. Since the average transit time between SST_ICW and SST_A was found to be ∼23 h and closely matched the 24-h period of intensification used in this analysis (vs. ∼13-h transit time between SST_IC and SST_A), only intensity change events that also included corresponding ΔSST_ICW values were used. Results from 24 ΔTCI events are included in Table 4a. In addition to statistics on ΔTCI, Table 4a includes upper and lower 50th percentile summary statistics for many of the variables listed in Table 3 including SST_A, SST_IC, SST_ICW, ΔSST_IC, ΔSST_ICW, TC_LAT, TCI_WIND, and TC_SPEED. Table 4a also includes satellite-derived estimates for upper-ocean heat content ahead of the storm (Q_H) as well as the following synoptic-scale atmospheric parameters: 200-hPa level zonal wind (U₂₀₀), 200-hPa air temperature (T₂₀₀), 850–200-hPa wind shear (S_850-200), 200-hPa divergence (D₂₀₀), and 200-hPa eddy flux of angular momentum (E₂₀₀). These atmospheric parameters, in addition to the weekly averaged climatological SST (SST_A-CLIM), were obtained from the Statistical Hurricane Intensity Prediction Scheme (SHIPS) database (DeMaria and Kaplan 1994, 1999) and were averaged over a TC-centered area of 500 km radius. Using a standard Student's t-test, statistically significant differences between upper and lower 50th percentile means were found at the 95% level (or higher) for ΔSST_IC and ΔSST_ICW (Table 4a, bold values). For the storms included in this research, inner-core SST change is linked to TC intensity change. Statistically significant differences between upper and lower 50th percentile means were not found for any other variable shown in Table 4a. However, while the findings are not statistically significant, results illustrated in Table 4a suggest that quick-moving, low-latitude, relatively weak storms may be more likely to intensify when compared to slow-moving, high-latitude, strong systems. Similar (nonstatistically significant) trends were also found in Table 3 between SST change and storm latitude, storm speed, and initial intensity.

It has been well documented that the magnitude of atmospheric shear can dramatically impact the intensity of tropical systems (Gray 1968; Merrill 1988). However, for the sample of storms used in this research, the difference in shear (S_850-200) between the group that noticeably intensified (upper 50th percentile) and the group that did not (lower 50th percentile) was not found to be statistically significant. It is probable that a number of physical processes impacted the rate of intensity change for the 24 events included in Table 4a. Nevertheless, the results presented in this analysis strongly suggest that the magnitude of inner-core SST change can, under certain circumstances, significantly impact the physical processes controlling storm maintenance and intensity change.

8. Maximum potential intensity and TC intensity change

Similar to results illustrated in Table 4a for ΔSST_IC and ΔSST_ICW, statistically significant differences between upper and lower 50th percentile means for ΔMPI_IC and ΔMPI_ICW were found. (These results are to be expected since MPI is a function of SST.) Comparisons between upper and lower 50th percentile ΔMPI means (given in m s⁻¹) and observed changes in hurricane intensity between the two groups (also in m s⁻¹) were conducted. As was done previously in the ΔSST analysis, only ΔMPI_ICW values were used since mean 23-h TC transit times from SST_ICW to SST_A more closely approximate the 24-h period of intensity change utilized in this study (relative to the ∼13-h transit time between SST_IC and SST_A). The statistical summary presented in Table 4b illustrates that average MPI_ICW reductions resulting from ΔSST_ICW ranged from −4.1 m s⁻¹ (for the upper 50th percentile) to −9.7 m s⁻¹ (for the lower 50th percentile). These results demonstrate the capacity of the upper ocean to potentially limit the magnitude of TC intensification while highlighting the variable nature of this “braking” process. The 5.6 m s⁻¹ difference in ΔMPI found between the upper and lower sample means represents 46% of the total 12.2 m s⁻¹ difference in TC intensity change found between the upper and lower 50th percentile means shown in Table 4b (i.e., 14.5 m s⁻¹–2.3 m s⁻¹). These results suggest that the “brake” on TC intensification may have been more “readily engaged” for the group of storms that intensified the least (i.e., lower 50th percentile sample).

These results also suggest that ambient MPI estimates may not always give an accurate account of MPI conditions found within the important hurricane inner-core environment. Results illustrated in Table 4b show that MPI_IC values were on average relatively close to MPI_A values for the events that significantly intensified (i.e., the upper 50% sample). However, by simply looking at mean MPI_A, one might have expected to find no significant difference in intensification between the upper and lower 50% samples given the fact that average initial MPI_A values were quite similar for both groups. These findings highlight the importance of obtaining accurate observations of SST within the hurricane inner core. In many cases MPI_A does not give an accurate measure of how close a system is to its maximum intensity where it matters most, within the hurricane high-wind environment. This is a significant point since potential intensity (i.e., MPI minus initial storm intensity) is an important and proven predictor used in the SHIPS forecast model (DeMaria and Kaplan 1994, 1999).

9. Summary

It is well accepted in the operational and research communities that the upper ocean can have a significant impact on maintaining and/or modifying TC structure and intensity. However, exactly how and to what extent variations in upper-ocean thermal structure directly impact local convective tendencies, overall TC structure, and ultimately, changes in storm intensity is still not well understood. This is partly attributed to the fact that ongoing upper-ocean and atmospheric modeling efforts are still lacking in several key areas (e.g., insufficient horizontal/vertical grid resolution, flawed/incomplete parameterization schemes, incomplete/crude physical representations of the atmospheric and oceanic boundary layers, etc). A likely explanation as to why changes in upper-ocean thermal structure have never been directly and quantitatively linked to changes in storm intensity is due to a limited number of observations. Both from an oceanic and atmospheric standpoint, the inner core is the most difficult region to routinely and accurately observe within the hurricane environment. Issues such as nearly continuous cloud cover, 10–20-m ocean waves, and wind speeds in excess of 50 m s⁻¹ make this region difficult for in situ platforms to survive, dangerous to traverse/circumnavigate, and nearly impossible for remote satellites to fully document. As a result, high-resolution, accurate depictions of the atmospheric and oceanic boundary layers within the TC inner core were rare prior to 1997. Since 1997, however, the use of highly durable and accurate GPS dropwindsondes and available AXBTs has enabled NOAA's Hurricane Research Division to penetrate and observe the TC inner and outer core boundary layer environments in several Atlantic hurricanes. By using the AXBT data from these storms in conjunction with archived SST data obtained during several “TC–buoy encounters,” the storm-to-storm variability associated with differences in SST between the hurricane inner core and the ocean environment ahead of the storm have been well documented.

Results from this research suggest that differences between inner-core and ambient SST are significantly less than horizontal SST changes typically observed in the post storm, TC cold wake environment (i.e., ∼0°–2°C vs. 4°–5°C). This finding should prove useful to modeling efforts attempting to verify the critically important (but infrequently observed) inner-core SST/mixed layer temperature. Estimates of upper-ocean heat content, energy extracted by the storm, and energy utilization were made. Findings from this analysis suggest that under most conditions, the upper-ocean heat content is an order of magnitude or more greater than the energy extracted by the storm. Results also show that relatively modest changes in inner-core SST can dramatically alter air–sea fluxes within the high-wind inner-core storm environment. Initial estimates show that SST changes on the order of 1°C lead to surface enthalpy flux changes of 40% or more.

For the subset of observations used in this study, it was shown that the magnitude of SST change (inner core minus ambient) was statistically linked to changes in TC intensity. These results suggest that storms experiencing reduced levels of inner-core SST cooling likely experience an increase in surface enthalpy flux, and as a result, are more likely to intensify. Ambient SST and upper-ocean heat content ahead of the storm were not associated with observed changes in storm intensity. Besides SST change, no other variable was statistically linked to changes in storm intensity (for the 24-event sample used in this study). Tropical cyclone intensity change is a complex, nonlinear process that often involves several competing or synergistic factors. Nevertheless, the results presented in this research strongly suggest that the (often “unseen” and numerically unaccounted for) variability associated with SST cooling within the hurricane inner core can, under certain circumstances, significantly impact the physical processes controlling storm maintenance and intensity change.