COMPARISON OF EFFECTS OF VARIOUS TROPICAL STORMS ON THE VERTICAL TEMPERATURE STRUCTURE OF THE OCEAN USING PICTORIAL REPRESENTATION Wi 1 1 iam Revesz, Jr. * * United States Naval Postgraduate School THESIS COMPARISON OF EFFECTS OF VARIOUS TROPICAL STORMS ON THE VERTICAL TEMPERATURE STRUCTURE OF THE OCEAN USING PICTORIAL REPRESENTATION by William Revesz , J r Thes is Advisor: Dr. Dale F . Lei ppe r September 1971 T14251 Approved faon. pub tic nzlzaHiZ; distribution witimitzd. Comparison of Effects of Various Tropical Storms on the Vertical Temperature Structure of the Ocean Using Pictorial Representation by William Revesz, Jr. Lieutenant Commander. United States Navy B.S., University of Tulsa, 1 96 2 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN OCEANOGRAPHY from the NAVAL POSTGRADUATE SCHOOL September 1971 c.i ABSTRACT To make comparisons of the effects of tropical storms on the ocean's vertical temperature structure, temperature-depth c ross -sect i ons were constructed using bathythermograph data and data from published articles. Upwelling, downwelling and m i x i ng , caused by tropical storms in deep and shallow water, are analyzed and compared. For a slow-moving, intense and very intense tropical storm, upwelling, from a depth of kO to 65 meters, is observed within the radius of hurricane-force winds. Downwelling as much as 20 meters occurs from 45 to 110 nmi from the path of the storm. This compares favorably with the theoretical results of O'Brien and Reid. A fast-moving, intense tropical storm has a similar effect on the vertical temperature structure if the thermocline is shallow,and upwelling, of a lessor degree than that caused by a slower-moving storm, can occur from a depth of 35 meters within the radius of h u r r i cane- f o r ce winds. A very fast-moving, very intense tropical storm can cause up- welling from a depth of 30 meters if the thermocline is shallow TABLE OF CONTENTS I. INTRODUCTION 9 II. PROCESSING OF DATA 11 A. STEP ONE 11 B. STEP TWO 12 C. STEP THREE 13 D. STEP FOUR 14 III. ANALYSIS OF TROPICAL STORMS 17 A. CROSS-SECTION A-A1 18 B. CROSS-SECTION D-D' 19 IV. COMPARISON OF FIGURES 22 A. EFFECTS IN DEEP WATER 22 1. Upwelling 22 2. Downwelling 23 3. Mixing 2k k. Sea Surface Temperature 2k B. EFFECTS IN SHALLOW WATER 2k 1 . Upwel ling 2k 2. Downwelling 25 3. Mixing 25 k. Sea Surface Temperature 25 V. COMPARISON WITH THEORY 27 A. SLOW-MOVING, INTENSE AND VERY INTENSE TROPICAL STORM 30 B. FAST-MOVING, INTENSE TROPICAL STORM 31 C. VERY FAST-MOVING, VERY INTENSE TROPICAL STORM 32 VI. CONCLUSIONS 33 VII. RECOMMENDATIONS 3k APPENDIX A--Cruise Track Used by GUS III for AFTER Hilda (196*0 35 LIST OF REFERENCES 51 INITIAL DISTRIBUTION LIST 53 FORM DD H73 55 LIST OF TABLES Table I Characteristics of the Tropical Storms, the Results of Analysis, and the Predictions of Theory Page 50 LIST OF FIGURES Figure 1 8 9a 9b 9c 9d 10a Cruise Track Used by GUS III for AFTER Hi Ida (1964) An Example of the Construction of the Figures Used for Analysis Cruise Track Used by R/V ALAMINOS for UNDISTURBED Hilda (1964) and BEFORE Betsy (1965) in the Gulf of Mexico Cruise Track Used by R/V ALAMINOS for AFTER Betsy (1965), and Location of Cross-sections C-C" through G-G" Position of C ros s -sect i ons A- A" and B-B" across the Track of Betsy (1965) with Location of Rad i o- t ransm i t ted BT ■ s Cruise Track Used by R/V HIDALGO for Thermistor-Chain Tows of BEFORE and AFTER Carla ( 1 9 6 1 ) and Location of Cross-section A-A' Cruise Track used by R/V ALAMINOS for BEFORE and AFTER Inez (1 966) and Location of Cross-section A-A' Cruise Track Used by R/V ATLANTIS II for BEFORE and AFTER Shirley (1965) and Location of C ros s -sect i on A-A1 Temperature-Depth C ros s -sec t i on A-A', UNDISTURBED and AFTER Hilda (1964) Temperature-Depth C ros s -sec t i on B-B1, UNDISTURBED and AFTER Hilda (1964) Temperature-Depth C ros s - sect i on C-C1, UNDISTURBED and AFTER Hilda (1964) Tempe ra tu re- Dept h C ros s - sec t i on D-D1, BEFORE and AFTER Hilda (1964) Temperature-Depth C ros s - sec t i on A-A", BEFORE and AFTER Betsy (1965), Based on Rad i o- t ransm i t ted Data Page 35 36 37 38 39 40 41 42 43 43 44 44 45 Fl gure Page 10b Temperature-Depth Cross-section B-B", BEFORE and AFTER Betsy (1965), Based on Radio-transmitted Data ^5 11a Temperature-Depth Cross-section C-C", BEFORE and AFTER Betsy Cl 965) in the Gulf of Mexico ^6 lib Temperature-Depth Cross-section D-D", BEFORE and AFTER Betsy (1965) in the Gulf of Mexico Zj£ lie Temperature-Depth Cross-section E-E", BEFORE and AFTER Betsy (1965) in the Gulf of Mexico ^7 lid Tempe ra tu re- Dep th Cross-section F-F", BEFORE and AFTER Betsy ( 1 965 ) in the Gulf of Mexico 47 lie Tempe ra ture - Dept h Cross-section G-G", BEFORE and AFTER Betsy (1965) in the Gulf of Mexico 43 12 Temperature-Depth cross-section A -A", BEFORE and AFTER Shirley (1965) k8 13 Temper a tu re- Dep th C ros s -sect i on A-A1, BEFORE and AFTER Carla (1961), Based on Towed The rm i s te r-Cha i n Data.... l±q 1 *» Tempe ra tu re- Dep th C ros s - sec t i on A-A1, BEFORE and AFTER Inez ( 1 966 ) ^9 ACKNOWLEDGEMENTS I would like to express my sincere appreciation to Dr. Dale F. Leipper for the patience, personal interest, and encouragement he provided in the preparation of this thesis. I would also like to thank Professor Ken Davidson for his encouragement and constructive criticism. A special thank you to my friend Patti Povemba , who spent many hours typing and retyping. INTRODUCTION Emphasis on the study of the effects on the sea surface and vertical temperature structure of the ocean by the pass- age of a tropical storm began in the mid 1950's. Evidence of marked cooling of the sea surface following the passage of tropical storms has been reported by several authors. These include Uda [19S4], Fisher [1958], Jordan [1964], Leipper [1967], Stevenson and Armstrong [1965], Landis and Leipper [1968], and Hazelworth [1968] . The first systematic observational studies on the three-dimensional effect of a hurricane on the sea sur- face and vertical temperature structure were conducted by Leipper on the effects of hurricane Hilda, (1964). Further studies were conducted by Landis and Leipper, and Franceschini and El-Sayed on hurricanes Betsy (1965) and Inez (1966), respectively. A theory for oceanic changes due to a stationary or slow-moving hurricane was developed by O'Brien and Reid [196?]; and O'Brien [1967, 1968]. Although somewhat limited by assumptions, the theory provides an explanation of some of the time-dependent non-linear processes involved. Basically, the initial ocean response to a hurricane is a development of an Ekman type flow which, in the Northern Hemisphere, leads to a diver- gence of the ocean surface layers away from the storm. The outward flowing warm water converges near the outer edges of the wind circulation and leads, through downwelling, to the formation of a deep, still warm, well-mixed layer in the ocean The mechanical effect of the wind and the convection brought about by heat loss from the sea surface to the atmosphere causes the mixing which occurs in the outward moving water. Near a coastline, the wind-induced motions are modified by the presence of a solid boundary, Franceschini and El-Sayed [1968]. To date, only Fisher [1958] and Hazelworth [1968] have concerned themselves with a comparison of the effect of more than one hurricane or typhoon. Both used sea surface temper- ature (SST) as the basic comparison parameter. The objective of this paper is to analyze and compare by use of selected numerical indices and pictorial representation, the effects on the vertical temperature structure of the ocean produced by hurricanes Carla (1961), Hilda (1964), Betsy (1965), Inez (1966) and typhoon Shirley (1965). 10 I I . PROCESSING OF DATA Leipper [1967], in his study of hurricane Hilda (1964), made observations and displayed vertical sections perpendicular to the hurricane's path. Comparison with data collected before the storm and with UNDISTURBED cros s -sect i ons (obtained during the next hurricane season prior to the passage of any storms and along the same cruise track lines) showed signifi- cant d i ffe ren ces in temperature and salinity. These differences appeared to be the result of the storm's action on the ocean. The warm surface waters were displaced to either side of the hurricane's path and a core of cold water with a shallow therm- ocline appeared near the center of the wake, demonstrating active upwelling. The Hilda study and further studies by Leipper , prior to and following Betsy (1965), indicated an area of downwelling outside the central upwelling region. This investigation and pictorial comparison of the effect of various tropical storms on the vertical temperature struc- ture of the ocean was initiated using the observations of Leipper [ 1 9 6 7 ] . A method of pictorial representation to represent the areas of upwelling and downwelling of Hilda 0964), was selected as follows: A. STEP ONE The locations of all bathythermograph (BT) observations made AFTER Hilda and all BT observations made of the To be published. 1 1 UNDISTURBED condition were plotted on the same chart of the Gulf of Mexico. The track of Hilda was superimposed over this data. Cross-sectional lines, perpendicular to the hurricane's track, were drawn with respect to the ship's track. For Hurricane Hilda, the placement of the cross- sectional lines was simplified because the AFTER and UNDISTURBED cruise tracks were made normal to the hurricane's track for both the first (AFTER ) and second (UNDISTURBED) cruises. In the case of Betsy in the Gulf , the BEFORE and AFTER plots of BT stations were overlaid. Three of the AFTER lines had been made exactly along the BEFORE lines. The other two c ros s -sec t i ona 1 lines were drawn normal to the path of the hurricane through the area of greatest observa- tional density except for c ros s -sec t i ons D-D" and E-E". All observations not falling directly on the cross- sectional lines were projected normal to the cross- sectional line. An example of the method of plotting the BT stations to the c ros s -sect i ona 1 lines is shown in Figure 1, which corresponds to the study of Hilda. B. STEP TWO To further refine the pictorial representation of the UNDISTURBED and AFTER vertical temperature structure, each Hurricane Betsy will hereafter be identified as Betsy in the Gulf or Betsy, the latter referring to hurricane Betsy in the Atlantic. 12 BT observation for the AFTER data was plotted with a vertical scale of 50 meters/inch. A horizontal scale of 60 nautical miles/inch was used to represent the distance from the path of the hurricane with the zero position representing the center of the hurricane path. Each BT observation was plotted at its respective distance from the center of the hurricane path and isotherms were connected between observations to form a temperature cross- section. A similar method was used to plot sections for the UNDISTURBED BT observations at their respective distances from the center of the hurricane's path. C. STEP THREE The plot of the UNDISTURBED isotherms was overlaid on the plot of the AFTER isotherms using the zero nautical mile position and the sea surface as the common points of allign- ment. With the vertical scales being the same, areas of upwelling, downwelling and mixing were readily discerned by comparing equal valued isotherms. To construct a pictorial representation of the effects of the hurricane for use in comparisons to other storms, the following procedure was used. With the AFTER plot overlaid on the UNDISTURBED plot, isotherms of the same value were compared starting from the surface. The lower stable isotherm was selected to serve as a reference for changes in the vertical temperature structure; that is, the isotherm was chosen as one which maintained the same depth within the area of upwelling both 13 In the AFTER and UNDISTURBED plot. Examination of the isotherms above this chosen lower reference isotherm showed that an isotherm for a 2°C greater temperature would adequately aid in representing depth changes due to the processes involved. In the case of Hilda, the lower and upper isotherms were 23°C and 25°C, respectively. Examination of the UNDISTURBED plot of isotherms indicated that the layer of water contained between these two isotherms, differing by 2°C, describes a layer with generally uniform thickness in the undisturbed water. In the case of Betsy (in the Gulf of Mexico) this index layer was best represented by the layer contained between the 26°C and 28°C isotherms. Henceforth, such layers contained between the two chosen isotherms will be referred to as the index layer. D. STEP FOUR The final construction of the figures used for Hilda and Betsy for pictorial representation was accomplished in the following manner. Dashed lines ( ) were chosen to represent the UNDISTURBED or BEFORE condition. Solid lines ( ) were chosen to represent the AFTER condition. When additional isotherms were needed to more clearly show the processes involved, a dash-dot line (-.-.) was chosen to represent the BEFORE or UNDISTURBED condition and a dotted line C ) was chosen to represent the AFTER condition. Large dots occurring in any of the above lines \k represent the location of BT stations as measured horizontally from the path of the hurricane and the depth of the isotherm as measured vertically from the surface. An example of this representation appears in Figure 2. The completed figures were then labeled and they represent simplified views of the temperature-depth c ros s -sec t i on s chosen for analysis and comparison. To construct figures for analysis of tropical storms Betsy (1965), Shirley (1965), Carla (1961), and Inez (1966, results representing the BEFORE and AFTER isothermal structure were extracted from published articles by Landis and Leipper [1968], Wright [ 1 9 6 9 J , Stevenson and Armstrong [1.965], and Franceshini and El-Sayed [1968]. A search of literature showed that these articles contained suitable results for comparison. These results were presented In figures similar to those constructed in this study. The figures were modified so that the depth scale would be one inch per 50 meters and correspond to the depth scale used in the Hilda and Betsy (in the Gulf) analyses. The horizontal distance scale assumed the dimensions per inch created by the one to one enlargement of the published data. This enabled a direct comparison of changes in the thickness caused by upwelling, downwelling and mixing of the index layer. After the modification of the published figures, steps three and four were completed and the figures were used to compare the depth changes involved. To facilitate 15 ease in comparison, all depths are presented in meters, distances in nautical miles, and temperatures in degrees Celsius. The final pictorial representations used for comparison were not meant to give exact values of horizontal measurement such as the exact distances where upwelling or downwelling occurs on either side of the hurricane's path. The isotherms chosen were representative isotherms and show approximate distances to the most dominant features The vertical pictorial representation indicates the change in depth and thickness of the index layer due to the effect of the hurricane on the vertical temperature structure. The isotherms for Hilda, and Betsy in the Gulf of Mexico, were plotted within + one meter of their true depth on the BT graph or on the value on the NODC data printout. There- fore, the vertical scale represents the actual change in depth and thickness of the index layer. The accuracy of the vertical scale of Betsy , Carla, Shirley and Inez, was assumed to be the same since original data was not used to plot i sothe rms . Section III explains in detail the construction of each temperature-depth c ros s - sec t i on . Analysis of a sample cross t section is included. Section IV is a comparison of the figures, and Section Visa comparison of the results of the pictorial analysis with the theoretical model proposed by O'Brien [1968]. 16 III. ANALYSIS OF TROPICAL STORMS Figures 1, 3 and k show the cruise track, cross- sections, and the extent of hurricane force winds of hurricanes Hilda [1964) and Betsy (1965) in the Gulf of Mexico. Figures 5, 6, 7 and 8 show the same information for tropical storms Betsy (1965), Shirley (1965),Carla (1961), and Inez (1966), respectively. Figures 9 through 14 are tempe ra t u re -dept h c ros s -sec t i ons of each tropical storm. Each cross-section was constructed normal to the path of that storm with the exception of c ros s - sec t i on s D-D" and E-E" of Betsy in the Gulf and A-A' of Shirley. Each temperature-depth c ros s - sec t i on was analyzed using the same procedures. The following is a sample analysis of Hilda (1964). The analysis examines two situations; first, a hurricane's effect on the vertical temperature structure in deep water, (c ros s - sec t i on A-A'); and second, the effect on the vertical temperature structure in shallow water (c ros s - sec t i on D-D1): The path of hurricane Hilda as it crossed the Gulf of Mexico from 30 September to 4 October 1964, is shown in Figure 1. When the hurricane was centered 250 nmi off shore in waters greater than 1,000 fathoms, she became more intense with winds up to 130 knots. The wind decreased to 105 knots as she moved toward the coast. The width of the zone having winds of hurricane force is shown in Figure 1. The average propagation speed was six to eight knots, and the width of the eye was approximately 35 nmi in the northern Gulf. 17 The data for the analysis was gathered in three ways. Some BEFORE data was collected by RV/ALAMINOS as it proceeded into port just ahead of the hurricane. Three bathy thermograph (BT) observations were made near the locations of BT ■ s 26, 21, and 18, respectively, as shown in Figure 1. Three additional BEFORE BT observations were obtained from the Bureau of Commercial Fisheries at locations near those indicated for BT's 57. 63, and 65 as shown in Figure 1. The remaining data was collected on cruise 65~A-11, conducted by the RV/ALAMINOS from 1 0 - 2 ^ August 1 9 6 5 and represents the UNDISTURBED condition taken eleven months later. Data was taken over the same paths used for observations of conditions after Hilda. The AFTER data was gathered using the 90-foot shrimp boat, GUS lil, operated by the Galveston Biological Laboratory, Bureau of Commercial Fisheries. The observations after the storm were made using the same BT instruments used on the RV/ALAMINOS. A. CROSS-SECTION A-A' (See Figure 1 for Location) The UNDISTURBED and AFTER data are plotted in Figure 3a, with all stations projected to the baseline A-A', constructed normal to and centered around Hilda's path. UNDISTURBED data was selected for analysis from hydrographic stations two through eight and NODC data printouts of BT's kO through 69, gathered on cruise 65-A-ll. AFTER data was selected from photographs of BT's 25 through 41, gathered 5 days after HMda by GUS III. 18 The UNDISTURBED condition was represented by the index layer contained between the 23 'C and 25 °C isotherms. The thickness of the index layer, 60 nmi either side of the path of Hilda, indicated a uniform, undisturbed layer approximately 10 meters thick, with the uppermost part of the index layer being 50 meters below the surface. The AFTER condition was represented by the same pair of isotherms. As a result of the radial divergence of the sur- face water, colder water was upwelled from the index layer to the surface for a distance of k$ nmi to the left and 50 nmi to the right of Hilda's path as shown by the surfacing of the 25°C isotherm. A reduction of SST by 3°C in this area was observed by comparing the UNDISTURBED and AFTER BT readings. Beyond 60 nmi to the left and right of Hilda's path there were strong indications of downwelling, probably caused by con- vergence of the warm surface water displaced from the area of up welling. In these areas the 25 °C isotherm was depressed approximately 18 meters deeper than the observed UNDISTURBED position. At a distance greater than 120 nmi to the right of Hilda's path, steeply sloping isotherms were observed and are believed to be caused by horizontal advection toward the hurricane path, Leipper [1965]. * B. CROSS-SECTION D-D1 The BEFORE and AFTER data are plotted in Figure 9d, with all stations projected to baseline D-D', which is normal to and centered to the far left of Hilda's path. BEFORE data was selected from BT ' s 10, 5 and 2, obtained by the Bureau 19 of Commercial Fisheries in 40 fathoms of water. AFTER data was selected from photographs of BT ' s 57 through 65, gathered nine days after Hilda by GUS III. This represents an area where Hilda reached shallow water and the wind-induced motions were modified by the presence of a coastline. The BEFORE condition was represented by the index layer defined by the 23°C and 25°C isotherms. The thickness of the index layer from 120 nmi to the left and 30 nmi to the right of Hilda's path represents a uniform, undisturbed layer averaging 10 meters thick, with the uppermost part of the index layer being about kO meters below the surface. The AFTER condition was represented by the same pair of isotherms. As a result of the radial divergence of the sur- face water, colder water was upwelled from the index layer to the surface for a distance of 20 nmi to **0 nmi to the left of Hilda's path as shown by the surfacing of the 25°C isotherm. A reduction of SST in this area of 5°C was indicated by comparing BEFORE and AFTER BT ' s . Seventy nautical miles to the left of Hilda's path there were strong indications of downwelling, probably caused by convergence of warm surface water displaced from the area of upwelling. In this area, the 25°C isotherm is depressed approximately 10 meters deeper than the observed BEFORE position. From 110 nmi to the left and 10 nmi to the right of Hilda's path, large areas of mixing were indicated by the equal distribution of the index layer above and below the BEFORE data position. 20 The data obtained from the analysis of all of the temperature-depth cross-sections prepared in this way is summarized in Table 1. 21 IV. COMPARISON OF FIGURES The comparison of figures is considered in two parts: hurricane effects in deep water and hurricane effects in shallow water. In each part, comparisons are made of upwell- ing, downwelling, mixing, sea-surface temperature decrease, and the extent of winds of hurricane force. A. EFFECTS IN DEEP WATER: 1 . Upwe 1 1 ? ng As a result of radial divergence of the surface water in all directions, sub-surface water is usually upwelled to compensate for the initial loss at the surface. O'Brien [1970] shows that the upwelling is a result of the influence of the radial component of wind stress. The component of wind stress was viewed as a steady state component by O'Brien, which enabled him to adequately describe the ocean dynamics under the core of the hurricane. Table 1 shows the comparison of the various cyclonic disturbances. In each case, the area immediately beneath the path of the hurricane was upwelled from a depth of at least kO meters to the surface. This is demonstrated by the surfac- ing of the upper portion of the chosen index layer. Currents have also had an effect on the observed location of the upwelled water. This is demonstrated in cros s -sec t i ons C-C", D-D", and E-E" of Betsy (in the Gulf). In the upwelled area, the chosen lower isotherm has remained essentially stable. This 22 indicates that water is brought in from below the mixed sur- face layer probahly causing a decrease in thickness of the index layer in areas near the outer limits of the hurricane force winds. Most cross^sect ions studied exhibited upwelling beneath the path of the storm and were offset to the left of the path of the storm as shown in cross-sections A-A1, B-B', C-C, D-D' of Hilda; B-B", C-C", D-D", F-F" of Betsy Cin the Gulf); A-A' of Shirley; A-A' of Carla; and A-A1 of Inez. The upwelling effect is a lasting effect, with indications, in at least the Hilda case, that it may last for up to twenty days. 2 . Downwe 1 1 i ng As a result of the radial divergence of the surface waters from the storm centers, still warm water displaced from the upwelled area converges near the outer area of the wind stress component. Leipper in his study of Hilda, and Betsy (in the Gulf of Mexico) found that downwelling was signifi- cant outside the upwelled area. Table 1 shows that an area of downwelling occurs near the outer edge of hurricane force winds which would agree with the onset of downwelling as described in theory by O'Brien [1967]- The mechanisms involved in downwelling are not as simple as those occurring directly beneath the hurricane and downwelling cannot be treated with the simple dynamic steady-state model used for upwelling. A common effect in the downwelled areas was a decrease in the thickness of the index layer and an increase in the depth of the index layer. 23 3 . Mixing As a result of the radial divergence of the surface water from the upwelled area and the combined effect of wind stress and turbulence in the diverging surface water, areas of deep, well-mixed waters are often encountered inside the storm area, and usually in concurrence with areas of maximum downwelling. An area of mixing is easily recognized by equal displacement of the upper and lower isotherm, thickening the BEFORE or UNDISTRUBED index layer. Mixing was not a common occurrence in deep water, but will be discussed in greater detail under EFFECTS IN SHALLOW WATER. k . Sea Surface Temperature A decrease in sea surface temperature was observed in the wake of each hurricane, and has been noted by many authors. The amount of decrease of sea surface temperatures is dependent upon the conditions existing prior to passage, the speed of passage, and the intensity of the cyclonic disturbance. The total decrease in the SST can not always be attributed to the effect of upwelling, but may be the result of heat loss to the hurricane, if no upwelling occurred B. EFFECTS IN SHALLOW WATER: 1 . U pwe 1 1 i ng Three c ros s -sect i on s demonstrate that the decrease in ocean depth and the presence of a coastline apparently alter the effects of a hurricane. C ros s - sec t i on D-D' of Hilda; cross-section G-G" of Betsy (in the Gulf), 2k and cross-section A-A' of Inez each show similar features. Upwelling is mostly confined beneath or slightly offset to the left of the hurricane centers. Also, these areas exhibit extensive upwelling. 2 . Downwe 1 1 ? ng Downwelling, as a result of the divergence of the warm water from the upwelled area, is not as clearly defined as in the deep water case. The extent of downwelling is less, being only several meters, as compared to observed effects in deep water. Areas where extensive downwelling was noted, such as c ros s -sec t i on A-A' of Carla, were possibly associated with areas of convergence of an induced current with an already present current in an opposite direction. 3 . Mixing The mixing in the areas of upwelling and downwelling is more common in shallow water as evidenced by c ros s - sec t i ons D-D' of Hilda, G-G" of Betsy (in the Gulf) and A-A' of Inez. The extensive mixing exhibited in the case of Betsy is probably a result of the effect of the geographical coastal boundary and the effect of the outflow of the Mississippi River . * k . Sea Surface Temperature The presence of a coastal boundary and the added complication of the outflow of a major river makes SST analysis difficult. The outflow of the Mississippi River places a warm, less saline tongue of water over the area 25 where Betsy had immediately passed. This tongue of water soon masks any effects of Betsy on the SST. 26 V. COMPARISON WITH THEORY O'Brien and Reid [1967] developed a theoretical decryp- tion of upwelling induced in a stratified, rotating, two- layered ocean by momentum transfer from an intense stationary, ax i a 1 1 y-symme t r i c atmospheric vortex. The dynamic internal response of the ocean was assumed to be ax i a 1 1 y-symme t r i c . Transfer of momentum between the air and the sea and between the upper and lower layers was allowed. O'Brien and Reid found that the results predicted by the model agree qualitatively with the following observations taken in the Gulf of Mexico after hurricane Hilda, 1967. Intense upwelling was confined within the radius of hu r r i cane- force winds. The displaced warm, central waters produced some downwelling outside the upwelled region. The maximum upwelling occurred at approximately 16 nmi from the hurricane path, which is an expected response to the maximum value of surface wind stress. The displaced warmer waters accounted for downwelling and thickening of the upper layer between h$ and 100 nmi from the hurri- cane path. A shallow mixed layer less than 25 meters deep was observed along the hurricane path and a deeper mixed layer ^60-80 meters) along the edges of the c ros s - sec t i on. O'Brien [ 1 9 6 7 3 continued the study with a second model that included mixing. He found that the velocities of the two models are essentially the same and mixing had little influence on the dynamic response of the system but that the 27 dynamic response does influence mixing. Mixing tends to lower the temperature and increase the salinity of the surface layer of the ocean oyer a broad region. Comparing the theoretical results to Hilda observations, O'Brien noted that in observations the mixing and upwelling were not symmetric about the hurricane path. He concluded that this may be due to the observed asymmetry of the wind stress distributions in a moving cyclone- The latter was not incorporated in the model. In a later study, O'Brien [1968], two important limitations to the above models were relaxed. They were; first, a specifically defined tropical storm was used to drive the ocean; and the second, this tropical storm was constrained to be stationary and symmetric. He relaxed these limitations to some extent and varied the initial layer depth from 30-150 meters (previous model 100 meters), the radius of maximum winds from 5~110 nmi (previous model 16 nmi), and varying the speed of the storm from three to eight knots. The relationships used by O'Brien were: du ou du dh r dv dv dv dh r ^ — + u^ — + v — + g^— = — ru- at 3x dy y^y ^n + ahu ahv 0 aPa + Tx /Ph - Tx /ph :*P Tb / ph T1 /Ph d „ a ■ + v y v ay at ax ay (1) (2) (3) whe re p is the density of the ocean f is the Coriolis parameter g is the acceleration of gravity 28 °a is the pressure at a radius in the hurricane h is the thickness of surface layer of the ocean t is the independent temporal coordinate u is the radial velocity in the ocean v is the tangential velocity in the ocean e is the density contrast T^, "m are the wind stress components at sea surface x y T , T' are the components of internal shearing x y stress at interface between layers. The results we re : 1. Cold wakes associated with upwelling occur sooner if the tropical storm acts on a shallow layer than if it acts on a deep layer. 2. As the tropical storm intensifies, upwelling is enhanced . 3. As the tropical storm intensifies, the extent of maximum upwelling becomes independent of the effective thermocline depth. For weak tropical storms, the extent of upwelling is highly dependent on the undisturbed thermo- cline depth. k . A slowly-moving tropical storm, even though not too intense, would tend to produce upwelling in its wake. A fast-moving storm might not produce much upwelling if the thermocline is deep, simply because there is too much water to move in a short time and the rate of momentum transfer to the ocean would be too slow. O'Brien's [1968] model for the response of the ocean to a slow-moving tropical storm was used for comparison with the data contained in Table 1, taken from Figures 9 through \k. Actual comparison with theory can only be made with slow moving, intense, and very intense tropical storms, when the speed of propagation falls within the three to eight knot range used in the development of theory. However, 29 conclusions reached by O'Brien can be extended to apply to faster propagating and more intense tropical storms,, since speed and intensity were independent variables, and varia- tions of these showed some specific tendencies; that is, that a tropical storm's increase in propagation speed lessens the amount and extent of upwelling, but an increase in intensity increases the amount and extent of upwelling. The comparison between results from theoretical model and conclusions drawn from observations of the present study is considered in three parts: first: slow-moving, intense tropical storms (Hilda, Betsy), and a very intense tropical storm, (Carla). Second: a fast-moving, intense tropical storm (Betsy in the Gulf); and third: very fast-moving, very intense tropical storm (Shirley). Observations from tropical storm Inez and c ros s - sec t i ons D-D' of Hilda and G-G" of Betsy (in the Gulf), were not compared to results from theory, since the nearness of the coastline and the shallow depth of water imposed boundary conditions not used i n theory . A. SLOW-MOVING, INTENSE AND VERY INTENSE TROPICAL STORM O'Brien and Reid and O'Brien have shown from numerical studies that the upwelled isotherm may decrease in depth from its original position as much as 80 to 90 meters and that the downwelling isotherm may increase in depth as much as 10 to 20 meters. Maximum upwelling occurs within the area of hu r r i cane- force winds and maximum downwelling occurs k5 to 100 nmi from the path of the tropical storm. 30 Comparison of c ros s -sec t i ons A-A1, B-B' and C-C' of Hilda and cross section A-A" and B-B" of Betsy to show the expected variations. As the intensity of the winds of Hilda decreased, the depth to which upwelling occurred (Sec. C-C1) also decreased. An examination of c ros s -sect ion A-A1 of Carla shows that the greater intensity of this hurricane resulted in upwelling from a deeper depth than in Hilda. This is as intuition and theory would predict. B. FAST-MOVING, INTENSE TROPICAL STORM Theory predicts that as the propagation speed increases, the amount and extent of upwelling and downwelling decreases for a given thermocline depth. Theory also predicts that as the depth of thermocline decreases, the amount and the extent of upwelling increases for a given speed of propagation. Hurricane Betsy in the Gulf of Mexico was of the same intensity as Hilda, but the depth of the thermocline was shallower. Examination of c ros s - sec t i on s C-C", D-D", E-E", and F-F", shows extensive upwelling and downwelling, but for a shallower depth than that of slower-moving hurricanes Hilda and Carla. This agrees with what theory predicts. Compari- son of the extent of upwelling and downwelling with that predicted for a slow-moving tropical storm of the same intensity shows agreement in c ros s - sec t i on s D-D" and F-F". C ros s -sec t i on C-C" and E-E" do not show this agreement, and is possibly due to the effect of the eddy described by Wunderly [1970] . 31 C. VERY FAST-MOVING, VERY INTENSE TROPICAL STORM This case has not been examined theoretically, but conclusions from theory of slower and less intense storms and observations imply that even with a very high speed of propagation, the high intensity of typhoon winds would cause a significant influence on the vertical temperature structure. C ros s -sec t i on A-A' of Shirley shows that Shirley, a storm of these characteristics, did affect the vertical temperature structure and caused water to be upwelled from 30 meters to the surface. 32 VI . CONCLUSIONS Previous studies of upwelling, downwelling and mixing caused by a tropical storm indicate the effects to be governed by several parameters. They are: 1. Initial depth of the thermocline 2. Propagation speed of the hurricane 3. Intensity of the hurricane k . Depth of water 5. Nearness of shallow water and coastal boundaries Comparison of tropical storms Hilda, Betsy, Carla, Inez and Shirley show that: 1. Observational data for slow-moving, intense trop- ical storms is relatively consistent from storm to storm, and agrees qualitatively with the theory of O'Brien, and O'Brien and Re i d . 2. Observations show that an increase in propagation speed, while maintaining the same intensity, results in less upwe 1 1 i ng . 3- Observations show that a decrease in intensity, while maintaining the same propagation speed, decreases upwe 1 1 i ng . k. Observations show that the presence of a coastal boundary and the shallow depth of water results in exten- sive mixing. This lends support to the validity of the assumptions of Franceschini and El-Sayed concerning hurri cane Inez. 5. The simple-index layer method of comparison is a useful method for comparing the effects on the vertical temperature structure caused by a tropical storm. 6. A search of literature shows that observational data suitable for this type of study is extremely scarce considering the number of hurricanes observed each year. 33 VII. RECOMMENDATI ONS It is recommended that: 1. Procedures be set up to take aircraft expendable bathythermographs (XBT's), prior to, and after severe tropical storms. Tracks normal to the projected path of tropical storms could be predicted with a fair degree of confidence at times, and XBT drops made. The ready avail- ability of data gathered this way would simplify and enhance the study of the effects of tropical storms on the vertical temperature structure of the ocean. 2. A computer study based on the model proposed by O'Brien [1968] should be conducted, incorporating the effects of evaporation, precipitation, sensible and turbulent heat transfer, and radiation exchanges with the atmosphere. The propagation speed of the tropical storm should be extended beyond the present value of eight knots. 3A APPENDIX A CRUISE HILDA TRACK SECTIONS AND B.T NOS. GUS m, OCT, 1964 Fig. 1. Cruise track used by GUS Ml for AFTER HI Ida (1964) . This figure represents the method of selection of cross- sectional lines A-A' through D-D' and the method of projection 35 JLL 120 6T 1— 60 8T 0 ST" r- 60 NMi 120 50 X a. UJ O 100 150 REPRESENTS BEFORE OR UNDISTURBED ISOTHERMS ADDED FOR CLARITY BEFORE CC Ul 5 LU o 50 100 150 A0 NMi 1?0 «r " a - er — I — f REPRESENTS AFTER ISOTHERMS 'ADDED FOR CLARITY AFTER Fig. 2. An example of the construction of the figures used for analysis. 36 o oo Fig. 3. Cruise Track Used by R/V ALAMINOS for UNDISTURBED Hilda (1964) and BEFORE Betsy (1965) in the Gulf of Mexico, (after Leipper [1967] ) . 37 Fig. A. Cruise Track Used by R/V ALAMINOS for AFTER Betsy (1965), and Location of Cross-sections C-C" through G-G", (after Leipper [ 1968J ) . 38 z 2. lO o o CO m CN o CN to '£><> in o Z > X" CO / X >- a. ui Q 200 - DASHED LINE IS UNDIS Fig. 9b. Temperature-depth c ros s -sec t i on B- B ' , UND I S i URBED and AFTER H i l da (l 964) . (See Fig. 1 for location.) 43 Fig. 9c. Temperature-depth cross-section C -C ' , UND I STURBED and AFTER H i 1 da (1 964) . (See Fig. 1 for location.) SFC OS UJ - 50 UJ 2 100 MIXING PATH OF HILDA DASHED LINE IS BEFORE ISOTHERM Fig. 9d . Temperature-depth c ros s - sec t i on D-D', BEFORE and AFTER H i 1 da (1 964 ) . (See Fig. 1 for location.) kk SFC 60 i 0 —r 60 — 1— 120 N.Mi // A I 50 X 100 »- o. HI Q 150 y- 26°C ---26°C 24°C 24°C UPWELL !NG y' PATH OF BETSY DASHED LINE IS BEFORE ISOTHERM Fig. 10a. Temperature-depth cross-section A-A", BEFORE and AFTER Betsy(1965) based on radio transmitted data. (See Fig. 5 for 1 oca t i on . ) sfc£ 180 120 50 OS UJ —100 o. ui a 150 60 i 0 60 H.Mi DOWNWELLING UPYVELLING PAn BETSY DASHED LINE IS BEFORE ISOTHERM Fig. 10b. Temperature-depth c ros s -sec t i on B-B", BEFORE and AFTER Betsy(1965) based on radio transmitted data. (See Fig. 5 for 1 oca t i on . ) 45 SFC 50 UJ Ul X 100 X ►- UJ a 150 120 N. Mi'/ UPWELLIHG UPWELLING /" \ / \ / PATH OF BETSY \ \ / DASHED LINE IS BEFORE ISOTHERM Fig. lla. Temperature-depth c ros s -sec t i on C-C", BEFORE and AFTER Betsy(l965) in the Gulf of Mexico. (See Fig. k for l.oca t i on . ) 120 N. Mi DOWNWELUNG 28°C 28°C 46°C I) -T DASHED LINE IS BEFORE ISOTHERM Fig. II b. Temperature-depth cross-section D-D", BEFORE and AFTER Betsy(l965) in the Gulf of Mexico. (See Fig. k for l oca t ? on . ) kG SFC 50 2 .100 150 120 N.Mi c'' PATH OF BETSY DASHED LINE IS BEFORE ISOTHERM Fig. 11c. Temperature-depth cr os s -sec t i on E-E", BEFORE and AFTER Betsy(l965) in the Gulf of Mexico. (See Fig. k f or 1 oca t i on . ) SFC 50 d: uj UJ 2 100 Ui O 150 120 N.Mi " DOWNWFLLING 28°C DASHED LINE IS BEFORE ISOTHERM Fig. lid. Temperature-depth cross-section' F-F", BEFORE and AFTER Betsy(l965) in the Gulf of Mexico. (See Fig. *t for location.) hi SFC 50 cc Ui Ul 1100 ►- Q. UJ O 150 - 120M.jftl TzFcT MIXING MIXING UPWELLING PATH OF BETSY DASHED LINE IS BEFORE ISOTHERM Fig. lie. Temperature-depth c ros s - sec t i on G-G", BEFORE and AFTER Betsy(l965) in the Gulf of Mexico. (See Fig. h for location.) ctr _100 Ul Q 150 PATH OF SHIRLEY DASHED LINF IS BEFORE ISOTHERM ^_s Fig. 12. Terppera tu re-depth c ros s - sec t i on A-A1', BEFORE and AFTER Sh i r 1 ey (J 965 ) . (See Fig. 6 for location.) 1*8 SFC — 50 cc UJ Ul 2 100 o. Ul O 150 0 T 60 N.Mi 120A' DOWNWE LNNG -s26vC 26*fc -'^24 C PATH OF CARLA DASHED LINE IS BEFORE ISOTHERM Fig. 13. Temperature-depth c ros s - sec t i on A-A', BEFORE and AFTER Car 1 a (l'96l ) based on towed t her mister chain dat< (See Fig. 7 for location.) 60 NMi Fig. 1 ^ . Temperature-depth c ros s -sec t i on A-A', BEFORE and AFTER Inez(l966), (See pfg. 8 for location.) hS 01 STANCE FROM PATH TO MAX. DOWNWELL ING ■1 65 80 75 30 70 50 70 ' NOT OBS 60 NOT OBS NOT OBS 18 120 ON PATH 85 90 110 110 NOT OBS 20 50 50 60 60 1.5-100 1.5-100 OEPTH OF WATER (METERS) 1000 1000 loo- iooo 100 1000 1000 1000 iooo 1000 1000 200- 1000 500 200- 2000 200 INFINITE TIME DATA WAS GATHEREO (DAY) O 03 UN O ^~~ cni rN O — -3- _ UN r^ <-, ON . , CO UJ - - • ~ f ? £ J; V + + m un * + + ■+ « + + z ~* "+ * u. cccz^'^ ' ~ !C o OOO* UN un fM r* £i:_=5 7 . on r^ ao uj un f*- *r + + +■ >~* w 1 1 w V ' ' ' ' z EXTENT OF HURRI CANE WINDS (NMI) o 70 8f 70 80 80 85 80 80 85 85 85 85 85 90 90 90 90 90 90 90 90 90 UKN 90 90 60 60 5- 100 5-100 MAXIMUM FORCE OF WINOS (KNOTS) oooo o o o o o o un o o O ^(s,_o f^^^rvjrM — o un un _^ PROPAGATION SPEED OF HURRI CANE (KNOTS) — *. -r *■ -^ -* ^T O OO OO CO CO CO CO CO j-uj^J-JJ-I O <= < U- I uj o h- uT _1 tt < < t- u. uj u> Z O >- C — UJ < uj i- x r _i _j «r uj O uj > < h- uj 3 _J o o ^3 X UN — h* — UJ z ox .- 1- 4. 4 o 60 65 60 65 55 50 50 50 110 70 90 85 120 55 1.0 25 1.1. UO NO DATA 35 OUN3Z3IJ15313 => ^ "^ O U>00 UN 1/1 UI M l~« ^N J,^, UI UI UI ^ ""^ <". UI < UN UN UN O O UN UN UN O O UN ^ ^^ ° O CC(- UJ UN J UN -JT f\ .3" r^\ J- rsj ^T — — UN UNuj O 2J < ^ O O UN O UN UN UNO UN < O ° °° ^ O < UN UN -x -J vOUJJ-'-N^NO^N'^ r^ON UJ -JT UJ OO^OOOD UNUN0UN00O00 <°° ^ OO CHAN OF OOWNWELLING LEFT < c 3/1.5 15/70 30/50 20/95 8/1.0 20/65 10/35 50/65* NOT OBS 1.0/65 20/80 NOT OBS 30/65 50/1*5* 25/30 1.0/55 8/30 12/1.0 8/30 8/38 NO DATA 6/70 15/80 10/35 10/11.0 OECREASE/INCREASE UP TO 20 MAX 1 MUM SEA SURFACE TEMPERATURE CC) u> ui < oou»u> oooooo0 *-* ° U»UJ a ^. z z < o ui O — X • c v. a o ~ ^r un UJ — — . uj - - --vO" ------ ON - UJ - UJ - ONOuJU.c, — - < ~-- < — < > c CD - ro c < v. 0) o 4-> fD o Q. O !_ I— - 4-J s_ O 4- 0) O-C 1— 10 04- — O 4-J 00 tn .— C 1_ o 0) — 4-J 4J o o fD — S-XI fD 0) x: 1- OQ_ fD 50 1 . 2. LIST OF REFERENCES Uda, M., On the Variation of Water Temperature Due to the Passage of Typhoon, Misc. Papers No . %~, Tokyo University of Fisheries, pp. 297-298, 1954. Fisher, E. L., "Hurricanes and the Sea-surface Temperature Field," J . Me teoro 1 . , v. 15, pp 333, 1958. 328- 3. Jordan, C. L., "On the Influence of Tropical Cyclones on the Sea Surface Temperature Field," P roc . Symp . Trop . Meteor., New Zealand Meteor. Service, Wellington, pp. 614-622, 1964. 4. Leipper, D. F., "Observed Ocean Conditions and Hurricane Hilda, 1964." J . Atmos . Sci . , v. 24, pp. 182-196, 1967. 5. Stevenson, R.E., and R. S. Armstrong, "Heat Loss from the Waters of the Northwest Gulf of Mexico During Hurricane Carla." Geofis. Intern., v. 5, pp. 49-57, 1965. 6. Landis, Robert C. and D. F. Leipper, "Effects of Hurricane Betsy upon Atlantic Ocean Temperature, Based on Radio- Transmitted Data." J. Applied Meteorol., v. 7, pp. 554-562, 1968. 7. 8. Hazelworth, John B. "Water Temperature Variations Resulting from Hurricanes," " J. Geophys. Res. , v No. 16, pp. 5105-5123, 1968. 73, O'Brien, J. J., and R. 0. Reid, "The Non-Linear Response of a Two-Layer Baroclinic Ocean to a Stationary, Ax i a 1 1 y- Symme t r i c Hurricane: Part I. Upwelling Induced by Momentum Transfer." J . Atmos . Sc ? . , v. 24, pp. 197-207, 1967. 9. O'Brien, J. J., "The Non-Linear Response of a Two-Layer Baroclinic Ocean to a Stationary, Ax i a 1 1 y- Symme t r i c Hurricane: Part II. Upwelling and Mixing by Momentum Transfer." J . Atmos . Sci . , v. 24, pp. 208-215, 1967. 10. O'Brien, J. J., "The Response of the Ocean to a Slowly- Moving Cyclone." Annalen der Me teoro 1 og ? e , v. 4, pp. 60-66, 1968. 11. Francesch i n i , G . A , and Sayed Z. El-Sayed, "Effect of Hurricane Inez (1966) on the Hydrography and Productiv ity of the Western Gulf of Mexico," Deutsche Hyd rog raph i sche Zeitschrift , v. 5, pp. 193"202, 1 9 6 8 . 51 12. Wright, Redwood, "Temperature Structure Across the Kuroshio Before and After Typhoon Shirley," Te 1 1 us XXI v. 3, pp. 409-i*l 3 , 1969. 13. Wunderly, W. L., Indicated Geostrophic Velocities and Volume Transports, Central and Eastern Gulf of Mexico, Warmest and Coldest Months., Master's Thesis, United States Naval Postgraduate School, Monterey, 1970. 14. Leipper, D. F., Hydrographic Station Data, Gulf of Mex i co , Texas A&M University Department of Ocean- ography reports 6 8 - 1 3 T , 1 9 6 8 . 52 INITIAL D ISTRI BUTI ON L I ST 1. Defense Documentation Center Came ron Station Alexandria, Virginia 2 2 31^ 2. Department of Oceanography Naval Postgraduate School Monterey, California 939^0 3. Oceanog raphe r of the Navy The Madison Building 732 North Washington Street Alexandria, Virginia 223H k . Dr. Ned A. Ostenso Office of Naval Research Code ^80 D Ar 1 i ngton , Va . 222 1 7 5 . Dr. D. F. Leipper Department of Oceanography Naval Postgraduate School Monterey, California 939 -iO 6. Professor Ken Davidson Department of Oceanography Naval Postgraduate School Monterey, California 939^0 7. Professor Robert 0. Re i d Department of Oceanography Texas ASM University College Station, Texas 778^3 8. Professor James J. O'Brien Department of Meteorology and Oceanography Florida State University Tallahassee, Florida 32306 9. Li brary , Code 021 2 Naval Postgraduate School Monterey, California 939^0 1 0 . LCDR William Revesz , Jr . 306 Genesee Street Trenton, New Jersey 08611 No . Cop i es 2 53 11. Commanding Officer Fleet Numerical Weather Central Naval Postgraduate School Monterey, California 939^0 12. Graduate Department Scripps Institution of Oceanography Box 109 LaJolla, California 92037 13- Dr. C. L. Jordan Department of Meteorology Florida State University Tallahassee, Florida 32306 14. Mr . I . Per 1 roth National Oceanog raph i c Data Center Washington, D.C. 20390 15. Mr. Henry Odom National Oceanog raph i c Data Center Wash i ngton , D.C. 20390 16. Dr . R. Cec i 1 Gentry National Hurricane Research Laboratory Box 8265 Coral Gables, Florida 33124 17. Dr. H. Burr Stei nbach Dean of Graduate Studies Woods Hole Oceanog ra ph i c Institution Woods Hole, Massachusetts 0 2 5 ^.3 18. Dr . R. Wr i ght Woods Hole Oceanog ra ph i c Institution Woods Hole, Massachusetts 025^3 19. Dr. G. A. Granceschini Department of Oceanography Texas A&M University College Station, Texas 7 7 8 ^ 3 l 20. Dr. R. F. Stevenson Scripps Institution of Oceanography Box 109 LaJolla, California 92037 21. Department of Physics The University of Tulsa 600 South Col 1 ege Tulsa, Okl ahoma , 7^104 No . Copies 1 5* Security Classification DOCUMENT CONTROL DATA -R&D [Security classification of title, body of abstract and indexing annotation must be entered when the overall report is classified) Originating activity (Corporate author) Naval Postgraduate School Monterey, California 939^0 2a. REPORT SECURITY CLASblFICATIOf Unclassified 2b. GROUP I REPORT TITLE Comparison of Effects of Various Tropical Storms on the Vertical Temperature Structure of the Ocean Using Pictorial Representation * DESCRIPTIVE NOTES (Type of report and,inclusive dates) Master's Thesis; September 1971 5 *u tmORiSI (First name, middle initial, last name) William Revesz , Jr. . REPOR T DATE Septembe r 1971 7m. TOTAL NO. OF PAGES 57 76. NO. OF RE FS »«. CONTRACT OR GRANT NO. 6. PROJEC T NO 9a. ORIGINATOR'S REPORT NUMBER(S) 9b. OTHER REPORT NO(S) (Any other numbers that may be assigned this report) 10 DISTRIBUTION STATEMENT This document has been approved for public release and sale; Its distribution is unlimited. II SUPPLEMENTARY NOTES 12. SPONSO RING Ml LI T AR Y ACTIVITY Naval Postgraduate School Monterey, California 939^0 3. ABSTRACT To make comparisons of the effects of tropical storms on the ocean's vertical temperature structure, temperature-depth cross-sections were constructed using bathythermograph data and data from published articles. y t rop i ca 1 nd compa red . i ca 1 storm, bserved within g as much as h of the storm, u 1 ts of al storm has ucture if the r degree than f rom a depth e winds. A n cause up- line is shall ow Upwe 1 1 i ng , d ownwe 1 1 i ng and m i x i ng , caused b s torms in deep an d shal 1 ow water, are analyzed a For a si ow-mov i ng , intense and very intense trop upwe 1 1 i ng , f rom a depth of hO to 65 meters, is o the radius of h urricane- force winds. Downwe 1 1 i n 20 me te r s occurs from kS to 110 nm i from the pat Th i s compa res f a vorab 1 y with the the ore t i ca 1 res O'Brien and Re i d . A fas t-mov i ng , in tense t rop i c a similar ef feet on the vertical tern perature str the rmoc line is sh allow, and upwelling , of a 1 es so that caused by a s 1 owe r- mov i ng storm , can occur of 35 meters wi th i n the radius of hu r r i cane- for c ve ry fa s t-mov i n g, very i ntense tropi cal storm ca we 1 1 i ng f rom a de pth of 30 meters if the thermoc FORM 1 NO V 68 /N 0101 -807-681 1 1473 (PAGE 1) 55 Security Classification A-3M08 Security Classification KEV WORDS Hurricanes Typhoon U pwe 1 1 i ng Air-Sea Interactions FORM 1 NOV 69 1473 (BACK 101-507-682 1 56 Security Classification IS ;t»tf 130324 Thesis I R365 Revesz Qf effects . compans troo-,Cal °f Var'°n the vertical storms on the i ' temperature struct _ * the ocean usany^K- "t-gjfii represented. 25700 ' 4 Thesis ^.30324 R365 Revesz c.l Comparison of effects of various tropical storms on the vertical temperature structure of the ocean using pic- torial representation. thesR365 Comparison of effects of < 3 2768 002 01322 9 DUDLEY KNOX LIBRARY '*