FOG ON THE CENTRAL CALIFORNIA COAST FOR 1973: ANALYSIS OF TRENDS John William Beards le> OUiUEY KNOX LIBRARY NAVAL POSTGRADUATE SCHOOL MONTEREY. CALIFORNIA 93940 NPS-58LR76031 NAVAL POSTGRADUATE SCHOOL Monterey, California THESIS FOG ON THE CENTRAL CALIFORNIA COAST FOR 1973: ANALYSIS OF TRENDS by John William Beardsley March 1976 Advisor: D. F Leipper Approved for public release; distribution unlimited. Prepared for: Naval Air Systems Command Washington D.C. 20360 T173138 UNCLASSIFIED SECURITY CLASSIFICATION OF THIS PACE (Whtr, Data Bntmrad) REPORT DOCUMENTATION PAGE READ INSTRUCTIONS BEFORE COMPLETING FORM t. REPORT NUMBER NPS-58LR76031 2. GOVT ACCESSION NO J. RECIPIENT'S CATALOG NUMBER 4. TITLE (and Submit) Fog on the Central California Coast for 1973, Analysis of Trends S. TYPE OF REPORT * PERIOO COVEREO Master's Thesis March 1976 «• PERFORMING ORG. REPORT NUMBER 7. AUTHORS*; John W. Beardsley In conjunction with Dale F. Leipper « CONTRACT OR GRANT NOMBERfaj 9. PERFORMING ORGANIZATION NAME ANO AOORE5S Naval Postgraduate School Monterey, California 93940 10. PROGRAM ELEMENT. PROJECT, TASK AREA * WORK UNIT NUMBERS II. CONTROLLING OFFICE NAME AND ADDRESS Naval Air Systems Command, 370C Washington, D. C. 20360 12. REPORT OATE March 1976 O. NUMBER OF PAGES 14. MONITORING AGENCY NAME * AOORESSO* ditfrfnt from Controlling Ofllea) Naval Postgraduate School Monterey, California 93940 IS. SECURITY CLASS, (oi thla ripon) Unclassified Tim. OECLASSIFI CATION/ DOWN GRADING SCHEDULE 16. DISTRIBUTION STATEMENT (ol thla Raport) Approved for public release; distribution unlimited 17. DISTRIBUTION STATEMENT (ol tha abalract antarod In Block 30, II dllloront from Rmport) IS. SUPPLEMENTARY NOTES 19. KEY WORDS (Conllnuo on roworaa mid* II nacaaaarr and Idrnntlty or block numbor) Fog, Forecasting, Inversion, Marine Fog, Synoptic, Visibility, California weather. 20. ABSTRACT (Conllnuo on rararao aldo II nacaaaarr and Idontltr by block mambat) Surface visibility data for selected stations on the central California coast in 1973 are analyzed. Radiosonde data from Oakland for the same period are used to derive meteorological indices. The year is divided into fog-related seasons, summer and winter, based on fog occurrence on the coast; and the seasonal and daily fluctuations of the indices are examined. A fog develop- ment model for the summer is formulated and compared to actual fog cases. In the winter, with far fewer coastal fog observations, the frequent DO ,^:M7, 1473 (Page 1) EDITION OF 1 NOV «■ IS OBSOLETE S/N 0 102-014- 660 1 I I Unclassified StCURITY CLASSIFICATION OF THIS PAGE (Whan Data Mnlarod) Unclassified JlCUWlTV CLASSIFICATION OP THIS PiGEWm D#-(« Fntmrmd 20. (continued) occurrence of frontal passages prevents a standard development model from being formulated and compared. Four Oakland soundings are compared with four radiosondes taken at NPS Monterey, and the Oakland soundings are found to closely approximate coastal conditions on these days. DD Form 1473 TT , .-. , 1 Jan 73 Unclassified S/N 0102-014-6601 2 SECURITY CLASSIFICATION OF THIS P*GEr*l»." D.(. Enn«d) Fog on the Central California Coast for 1973 Analysis of Trends by John William Beardsley it Commander, United St B.S., United States Naval Academy, 1964 Lieutenant Commander, United States Navy Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN OCEANOGRAPHY from the NAVAL POSTGRADUATE SCHOOL March 1976 NAVAL POSTGRADUATE SCHOOL Monterey, California DUHLEY KNOX LIBRARY NAVAL POSTGRADUATE SCHOOL WONTEREy.cal,FORNU9^ Rear Admiral Isham Linder, USN Superintendent Jack R. Borsting Provost This thesis prepared in conjunction with research supported in part by NAVAL AIR SYSTEMS COMMAND under A370 370C/186B/ 6F52-551-700 issued 2 July 1975. Reproduction of all or part of this report is authorized. Released as a Technical Report by: ABSTRACT Surface visibility data for selected stations on the central California coast in 1973 are analyzed. Radiosonde data from Oakland for the same period are used to derive meteorological indices. The year is divided into fog-related seasons, summer and winter, based on fog occurrence on the coast; and the seasonal and daily fluctuations of the indices are examined. A fog development model for the summer is formulated and compared to actual fog cases. In the winter, with far fewer coastal fog observations, the frequent occurrence of frontal passages prevents a standard development model from being formulated and compared. Four Oakland soundings are compared with four radiosondes taken at NPS Monterey, and the Oakland soundings are found to closely / approximate coastal conditions on these days. TABLE OF CONTENTS I. INTRODUCTION 11 II. OBJECTIVES 13 III. BACKGROUND 14 A. RELIEF FEATURES 14 B. ATMOSPHERIC CIRCULATION 17 C. SEA-SURFACE TEMPERATURE 19 D. TEMPERATURE INVERSION 25 E. FOG FORMATION 26 F. FOG DISSIPATION 28 G. SUMMARY 29 IV. APPROACH 32 A. SELECTION OF DATA 32 / 0 B. TREATMENT OF DATA 36 1 . Annual Data 36 2 . Daily Data 39 V. DATA ANALYSIS 41 A. YEARLY ANALYSIS 41 B. DAILY ANALYSIS - 51 1 . General Comments 51 2. Summer 53 3. Winter 66 C. COMPARISON OF OAKLAND AND NPS RADIOSONDES 69 VI. CONCLUSIONS 73 VII. RECOMMENDATIONS 75 APPENDIX A - YEARLY DATA 77 APPENDIX B - DAILY DATA 82 REFERENCES 113 INITIAL DISTRIBUTION LIST 116 LIST OF TABLES TABLE I. Average Hourly Wind Movement at Point Reyes 18 II. Mean Air Temperature at Point Reyes 19 III. Daily Average Air Temperature at Alameda Naval Ai r Station 19 IV. Mean Surface Water Temperatures for Selected Central California Stations 25 V. Comparison of Sea-Surface Temperature Measurements for the Coastal Area off San Francisco 35 VI. 1973 Marine Fog Occurrence at Central Pacific Coastal Stations 41 VII. Percentage of Hours with Fog at Coastal Stations 42 VIII. Inversion Statistics - Oakland, 1973 44 /IX. Distribution of Cases in which the 0400 (Local) Oakland Radiosonde showed no Inversion to Exist 46 X. Inversion Statistics - Open Ocean off San Francisco 1973 47 XI. Comparison of Fog Parameters between Summer and Winter, 1973 50 LIST OF ILLUSTRATIONS FIGURE 1. Data Stations along the Central Pacific Coast 16 2. Mean Sea-Surface Temperature - Winter 20 3. Mean Sea-Surface Temperature - Spring 21 4. Mean Sea-Surface Temperature - Summer 22 5. Mean Sea-Surface Temperature - Fall 23 6. Sea-Surface Temperature for Fog Case of 28 September 1973, Monterey Bay 65 7. Upper-Air Temperature - Oakland and Monterey, July 1973 71 TABLE OF SYMBOLS AND ABBREVIATIONS BI Base of the temperature inversion FAA Federal Aviation Agency MI Moisture Index NAS Naval Air Station NOAA National Oceanic and Atmospheric Administration NPS Naval Postgraduate School T Surface air temperature T Sea- surface temperature T Temperature at the top of the temperature inversion TI Temperature Index UI Upwelling Index --* ACKNOWLEDGEMENT The author wishes to express his sincere appreciation to his thesis advisor, Dr. Dale F. Leipper of the Naval Postgraduate School, for his patient guidance, advice and encouragement in this study. Gratitude is also expressed to Professor Robert J. Renard and Professor Glenn H. Jung, whose constructive comments aided greatly in the completion of this paper. In particular, the author wishes to thank his wife, without whose understanding and assistance this undertaking could not have been accomplished. 10 I. INTRODUCTION A great deal of data and documentation is available on advection fog for the central west coast of North America. However, on closer examination of the available references, a gap appears in the sequence of studies carried out to date. Byers [1930] and Patton [1956] conducted extensive investigations of central California summer marine fog on the coast and in the San Francisco Bay area. McClure [1974] examined the relation between temperature inversions and marine fog, based on ten months of data, but he used Oakland as his only central California station rather than an actual coastal station. Peterson's and Leipper's 1975 study extended Leipper's 1948 fog prediction model to cover the central California coast during the period May through September. Most of the effort which has gone into the above papers has been limited to the summer months. Therefore, considering the well-documented hazards imposed by marine fog to military and civilian operations and also the deficiency of information on fog development for the winter season on the central California coast, it was determined that a study of this nature was appropriate and should be undertaken. Specifically, a statement of the problem is to conduct an examina- tion of air mass fog formation on the central California coast, utilizing a full year of data. 11 In this study, upper-air radiosonde data will be plotted with visibil. ity data from coastal stations. Non-diurnal indices will be computed, and these data will be used to examine actual fog sequences and deter- mine a sequential model. This model will be compared to actual fog developments for 1973. ,* 12 II. OBJECTIVES A. To examine one full year of meteorological data on the central California coast, in order to determine which parameters could be used as fog forecasting aids. B. To observe the seasonal and daily variation of the parameters in A. C. To examine the individual cases of fog, in order to formulate a standard fog development model for each season. D. To estimate the accuracy of the Oakland radiosonde in reflecting upper-air conditions over the central California coast. 13 III. BACKGROUND In order to study marine fog along a length of coastline one must consider the effect which topography has on various meteorological parameters. Also, the synoptic weather picture must be understood, and the interaction between the ocean and atmosphere examined. Only after each of these basic yet indispensable areas is understood can one proceed to study the formation and characteristics of marine fog in the area of interest. In this section, selected factors which affect coastal fog are reviewed: A. RELIEF FEATURES B. ATMOSPHERIC CIRCULATION C. SEA-SURFACE TEMPERATURE D. TEMPERATURE INVERSION E. FOG FORMATION / F. FOG DISSIPATION G. SUMMARY A. RELIEF FEATURES Topography has a major effect on local< fog and stratus. Coastal mountains tend to contain the atmospheric marine layer [Rosenthal 1972]. Schroeder, et al [1967] discuss the role of terrain features in controlling the interaction between the cool, moist marine air mass and the warm, dry continental air mass. They state that a steep, mountainous coast- line will restrict sea breeze penetration to only a few kilometers, whereas a flat, unobstructed coastline will allow penetration of up to 300 km. The coast of California between San Francisco and Monterey is oriented northwest-southeast. The continental shelf south of 14 San Francisco narrows to less than five miles wide and coastal mountains rise abruptly from most of the shoreline to heights of over 1220m (4000 ft) [Patton 195 6]. The Santa Cruz range forms the coastal chain, running from half-way down the San Francisco Peninsula to an area ten miles in- land from the northern shore of Monterey Bay. The Santa Lucia range commences at the southern shore of Monterey Bay and continues south- eastward along the coast at altitudes exceeding 1220m. Significant features of the coastal mountains are the gaps which allow marine air into the flat inland areas. In addition to the Golden Gate, there are three gaps in the coastal Santa Cruz mountain range between San Fran- cisco and Monterey Bay; the San Bruno Gap (61m elevation) and the higher and narrower Crystal Springs Gap, both in the Bay Area, and the Pajaro River Gap. The latter is the low wide valley along the eastern shore of Monterey Bay which separates the Santa Cruz and Santa Lucia mountains * and forms the mouth of the Salinas Valley. / Topography is an important feature which greatly affects the gen- eral atmospheric circulation of the region, as will be shown in the next section [Patton 1956]. Figure 1 shows the locations of the land stations from which data was collected for the present study. Station heights above mean sea level are: Oakland 1. 9 m Monterey Airport 49.2 m Pillar Point 40. 0 m Hidden Hills 262. 0 m Pigeon Point 9.2 m R/V Acania 2. 0 m Point Pinos 8.6 m Pillar Point and Pigeon Point are located on the coast and are direct- ly exposed to marine atmospheric conditions. Point Pinos is protected from southerly winds, but is exposed to prevailing northwesterly winds. 15 Rues OAKLAND fHKPOKT f\LUMt PfllMT ft«eoM Fb+*r F*'6uft.g L. Tag CeKfnstfe Z*L&o(Zti\* Cohzt ©*<£L«a*I 0 Al**>*T HioteH Ftom Ci 6 3 Omar 5020 feviseo l/iS/frf 16 B. ATMOSPHERIC CIRCULATION The circulation of air over the San Francisco Bay area is controlled by the position of the North Pacific anticyclone [Patton 1956]. In the winter months, the North Pacific high center lies to the southwest of California where it assumes the configuration of a less dominant high without the shape, gradient or circulation of a well defined anticyclone. In the summer, the anticyclone strengthens and it moves to the north- o o west, centering itself around 40 N, 140 W. Also developing in the summer over the central valley of California is an intense low pressure system extending from Mexico to Oregon, formed by the radiant heating of the flat terrain. This summertime low is usually separated from the North Pacific high by the inland mountain chain [Byers 1938, Patton L956]. / Together, the strengthened North Pacific anticyclone and the inland low pressure trough combine so that the isobars run approxi- mately parallel to the Pacific coast. The clockwise circulation around the high, along with surface friction, causes* the surface winds to blow from the northwesterly or westerly direction. These imposing pressure features completely override the diurnal land -sea-breeze system characteristic of many coastal areas. According to Byers [1938], surface wind from the land is very rare in the summer and usually results only in a decrease of the on-shore winds. However, drainage wind from the land is common by night at Monterey in the summer. [R. J. Renard, NPS, personal communication]. 17 The following table, based on 31 years of data, shows the domi- nance of the northwest surface winds, especially in summer: TABLE I Average Hourly Wind Movement at Point Reyes [from Patton 1956] Month J F M A M J J A S O N D Year Prevailing direction NW NW NW NW NW NW NW NW NW NW NW NW NW Velocity (km/hr) 2a2 31.4 33.7 3ao 42.4 43.4 36.2 32.2 29.3 27.5 26.7 2a3 33.0 When the North Pacific anticyclone moves eastward and its iso- baric patterns overlap the Pacific coast, the inland low pressure system is forced southward, resulting in a transport of continental air toward the San Francisco Bay area from the-east and producing high temperatures. If the anticyclone moves south, as it does in the winter, cyclonic storms may pass and produce rainfall and fog along the coast [Patton 1956]. Tables II and III illustrate the difference in monthly air tempera- tures between Point Reyes, which is directly impinged upon by the California current and atmospheric marine layer, and Oakland, which is partially shielded from direct ocean influence by the Santa Cruz and other mountains. The greatest difference occurs during the summer months as a result of the increased solar radiation inland, as well as the North Pacific high causing inland subsidence and heating of the upper air. 18 TABLE II Mean Air Temperature at Point Reyes (34 years) [from Byers 1938] Month J F M A M J J A S 0 N D Year Tempera- 9.7 9.9 10.2 10.3 10.7 11.3 12.0 12.5 13.6 13.0 12.1 10.4 11. 3 ture (C) TABLE III Daily Average Air Temperature at Alameda Naval Air Station [from NAS Alameda Forecasters Handbook, 1972] Month J F M A M J J A S O N D Tempera- 9.8 1L9 13.1 146 15.7 17.3 17.9 17.9 18.9 16.6 142 10.7 ture (C) In its normal summertime position, the Pacific anticyclone serves to insulate the Pacific coast from the intrusion of the few weak / summer low pressure systems [Patton 1956]. C. SEA-SURFACE TEMPERATURE Two factors cause the Central and Northern California coastal waters to be colder than normal during the summer and fall: 1) the California current and 2) coastal upwelling. According to Sverdrup et al [1942], the California current is a sluggish continuation of the Aleutian current, approximately 700 km wide which flows adjacent to the coast in a southeasterly direction. Figures 2-5 show the sea- surface temperature pattern resulting from the combined effects of 19 PeePHteo av Fueer We*™**. CGMrttAU QuumBA A/#v»t /hit StrmriotJ 20 Me** StfA $uAr»c0 T£/*ficm*TuA£ -# 21 ^ 22 Fall ( Octo8&j /UovemAen) ^ 23 these two mechanisms. Figure 4 shows a crescent of cold water cen- tered on Cape Mendocino. Byers [193 8] explains this anomaly as due to upwelling of colder water during the period April-August. Bakun [1974] however, shows conclusively, using seven years of data (1967- 1973), that upwelling at Cape Mendocino occurs from March to mid- September, and that as one moves further south to the tip of the Baja Peninsula, upwelling is prolonged for the entire year. By June- July, the cold sea surface area has spread south to San Francisco. In August, the sea-surface temperature off of San Francisco has reached a minimum, and from this period, the temp- erature rises until temperatures offshore and alongshore are nearly equalized in March [Patton 1956]. As will be shown, the upwelled -fongue of cold water plays an Indispensable role in influencing the formation of the atmospheric inversion, which in turn controls fog and stratus formation. Table IV shows the difference in sea-surface temperature between San Francisco, Alameda (on the continent side of San Francisco Bay) and Pacific Grove on the Monterey Peninsula. It illustrates graphically how much sea-surface temperatures differ at coastal locations having various degrees of exposure. 24 TABLE IV Mean Surface-Water Temperatures for Selected Central California Stations (C) [from NOAA, U. S. Coast Pilot 7, 1975] Month J F M A M J J A S O N D Mean San Francisco (Fort Point) 10.4 10.9 11.6 12.4 13.1 13.9 147 15.2 15.5 14 8 13.0 11.2 B.l Alameda Pacific Grove 10.3 11.8 11.9 12.0 13.9 12.2 16.1 12.4 17.8 12.8 19.4 13.4 20.5 13.8 20.5 13.9 20.2 142 17.7 13.7 144 12.9 11.4 12.4 16.2 13.0 D. TEMPERATURE INVERSION Leipper, in his 1948 paper on winter fog development in San Diego, stated that the first stage of fog development is the presence of warm air over colder water, which guarantees the formation of a surface temperature inversion. The inversion serves to restrict the vertical -movement of moisture and causes the thin lower layer to approach saturation [Leipper 1948]. A crucial question arises - where does the warm air come from? There have been a number of different explanations advanced. Byers [1930] stated that the warm air in the upper layer is a permanent condition, and the cooler marine layer simply flows underneath. Bowie [1928] identified two separate air masses, with a warm continen- tal air mass being on top. Leipper [1948], Patton [1956], Petterssen [193 8] and Schroeder [1967] offered the most valid explanation; namely, that the warm air is the result of upper air that has subsided and warmed 25 adiabatically. Leipper [1948] pinpointed the cause of subsidence by- specifying that the North Pacific high moves inland, resulting in a general easterly flow which progresses downslope along the mountains east of San Diego and is heated adiabatically. As the warm air comes in contact with the colder surface water a temperature inversion is formed. Patton [1956] contended that the subsiding air mass, more than any other factor, affects the height of the base of the inversion. The Forecasters Handbook for Naval Air Stations Alameda [1972] and Moffett Field [1974] acknowledged subsidence as the factor which warms the air and, combined with cooling from below, forms the inversion in the San Francisco area. E. FORMATION OF FOG Byers [1930] attributed the formation of advection fog at sea to / ' . the passage of saturated air over the colder upwelled coastal waters. Formation. is greatly facilitated by the presence of hygroscopic salt particles which act as nuclei around which fog droplets form. Petterssen [1938] refuted the contention that fog is formed by cooling from below. He proposed that the observed unstable conditions beneath the inversion were the result of convection and turbulent mixing . Patton [1956] agreed with Byers, but clarified his position: "Stratus cannot occur without cooling of the surface layers by contact with upwelling water, but the original temperature of the air mass 26 determines whether such cooling is sufficient to produce minimum temperatures at the base of the inversion low enough to cause condensation. " Leipper [1948] described the formation of fog in stages, the first of which is the advection of warm continental air over colder water with the consequent development of a temperature inversion. Following this, the dry air aloft permits increased radiational cooling of the moist sur- face layer and consequent initiation and/or intensification of the fog. Goodman [1975] in her examination of the microstructure of San Francisco Bay area fog, demonstrated that cooling from above by radiation was responsible for a 1-2 C night time drop in temperature in the marine layer. The Forecasters Handbook for Naval Air Station Moffett Field [2974] describes the two types of California fog as advection (as noted by Byers, Patton, and Leipper above) and radiation. Advection fog can form over land or water, but radiation fog, at least according to the Forecasters Handbook, can form only over' land when the earth, cooled by radiation, cools the air to its dewpoint. However, radiation fog can be advected for short distances over a body of water. By virtue of the almost constant northwesterly prevailing winds in the marine layer, advection fog appears to compose the major share of fog occurring on the exposed California coast. The moist marine air passing eastward over the water is several degrees warmer than 27 the cold tongue of upwelled surface water along the coast. This marine air causes the inversion base to lift off of the surface. Beneath the inversion, the marine air continues to cool, and as the dewpoint is passed, fog forms. The fog is then advected toward the shore by the prevailing winds. The top of the fog or stratus layer normally coincides with the base of the inversion [Byers 1938, Patton 1956]. F. DISSIPATION OF FOG Dissipation of fog is caused by heating of the cool, moist marine layer, so that the air temperature is raised above the dewpoint, causing evaporation of the fog. Leipper [1948] observed that after marine fog forms at sea, it becomes thicker due to radiation cooling from the top and mixing, thereby decreasing the air temperature to below that of the sea surface. Sea surface evaporation continues. As the fog layer initially invades the coastal area, it may be dissipated by the daytime heat of the land. However, eventually the fog layer becomes deep enough to maintain itself further inland over the warm land. In the San Francisco Bay area, Byers [1930] found that inflowing afternoon fog seldom passed beyond the Golden Gate, due to mixing with the warmer turbulent land air and subsequent evaporation. Only after the air inside the Golden Gate had cooled sufficiently, usually shortly before sunset, would the fog be able to maintain itself. Al- though Byers found that the maximum fog thickness may be in excess 28 of 600 m, he observed that the average thickness was about 400 m. Leipper, in his 1948 San Diego study, recorded a maximum thickness of 400 m. Byers [1938] also noted that as fog flowed over hills surround- ing the Bay area, adiabatic heating on the leeward slopes caused dissipa- tion about 100 m below the crest. Dissipation of fog also can be caused indirectly by the eastward movement of the North Pacific high. As it overlaps the northwestern states, the anti-cyclonic circulation initiates an easterly flow of warm, subsiding air which forces the inversion base nearly to the ground and along with mixing dissipates any fog or stratus which may exist. These occasions of inland movement of the high are known for extremely clear weather, usually lasting several days [Forecasters Handbook, Naval Air Station Alameda 1972]. . -" / G. SUMMARY From the discussion of the various parameters which affect west coast marine fog formation and movement, it is. seen that: 1. Relatively warm air, circulating over the cold tongue of water caused by coastal upwelling, is cooled and an inversion is created. 2. In the summer, when the North Pacific high has migrated to the northeast and a ridge extends over land, subsidence strengthens the inversion, and this very stable layer serves effectively to separate the cool moist marine air in the lower levels from the warm dry air at higher levels. 29 3. Beneath the inversion, the air over the areas of upwelled water continues to cool to the dewpoint, and condensation of water particles in the air occurs. This process could not exist without the presence of cloud condensation nuclei. 4. The return of the prevailing winds from the northwest drives the fog across the coast and against the slope of the coastal mountains. 5. If the fog is shallow and the inversion is low, dissipation will occur shortly after the fog crosses the coast. 6. The low pressure area over the central valley, caused by intensive heating, serves to reinforce the difference in pressure across the coastline and to draw the modified marine air inland through the mountain gaps. 7. The orientation of the mountains and valleys in the direction 6i flow of the prevailing wind channels the inflowing fog to the low areas inland of the coastal range. 8. The fog remains until heating from above and below causes evaporation. ,** ■ Crouch [1973] presents a graphic picture of the way in which he has observed marine fog crossing the Pajaro River- Monterey Bay gap and spreading to the inland valleys during the summer months: "As the heated air of the [Salinas] valley rises during the day, a layer of cool air from the ocean begins to move inland, carrying masses of fog into the valley of the Salinas and pressing 30 it against the Aromas hills and the low ridges near San Juan Bautista that mark the northern limits of the valley .... o Salinas may be a chilly 50 while 40 miles away in King City the thermometer will stand at 90 , and in the side valleys the heat will go up past the 100 mark Long exploratory streamers of fog begin to move slowly southward. The white tentacles become patches and the patches become billows, moving past Salinas and Spreckles and Chualar .... At the same time another stream of fog moves up Carmel Valley, borne inland by a steady wind. It climbs the ridges of the Aguajito and Jacks Peak, curling up over Los Laureles and down into Corral de Tierra. Slowly the fog piles higher against the western slopes of Mdunt Toro and Palo Escrito ridge until, bursting over the top in a graceful banner, it flows downward to join the white river moving silently up the Salinas Valley. " 31 IV. APPROACH A. SELECTION OF DATA After the initial decision was made to examine an annual fog cycle on the Central California coast, it was necessary to specify the loca- tion and time frame for the study. The coast between San Francisco and Monterey was chosen because of the close proximity to the Oakland International Airport, one of the very few locations where daily upper - air soundings were observed. Also, there was good coverage of this area provided by coastal stations at Pillar Point, Pigeon Point, Point Pinos and the Monterey Airport. The period 1973 was chosen because of data availability. Data provided by McConnell [1975] listed Oakland soundings from 1968 through i June 1973. The soundings for the remainder of 1973 were obtained from the National Weather Records Center through the Naval Weather Service Detachment, Asheville, North Carolina. Surface weather observations from Point Pinos and Pigeon Point were available commencing October 1972, but Pillar Point records were not available prior to January 1973. Additionally, FAA observations from Monterey Peninsula Airport were available for 1972 and later, as well as Hidden Hills temperature and relative humidity records. Also, surface weather observations from the deck log of R/V Acania were on file for 1973. Consequently, 32 1973 was chosen as the year with the most complete observations on hand. 1 The 0400 (PST) radiosonde was chosen over the 1600 sounding because 1) the continental air is best represented by the early morning flow which exists at 0400, 2) more fog occurred in the early morning hours than in the afternoon, thereby enabling more frequent observation of upper-air conditions during periods of fog, and 3) the 0400 sounding is the one used by forecasters for predicting the day's weather. Sea-surface temperature exerts a controlling influence on fog formation. A significant problem which arose in the selection of data was the choice of which sea-surface temperature to use for computation of the Temperature Index and Moisture Index. Data from four sources were available: 1) A supplement to Fishing Information, published by the National Marine Fisheries Services at 15 -day intervals, contained o sea- surface temperature charts contoured at 2 F, based on data from merchant vessels, fishing vessels, and airborne infra-red surveys; 2) A diagram from Bakun, et al [1974] showed sea-surface temperatures for the Pacific coast, based on 20 years of data; 3) A graph of monthly mean surface-water temperatures (and densities) for San Francisco (Fort Point) based on 5 1 years of data, appeared in NOAA United States Coast Pilot 7 [1975]; 4) Monthly mean sea-surface temperature measure- ments at Fort Point for 1973 were unpublished data obtained from Due to the important diurnal ramifications of fog development, all times are PST unless otherwise designated. 33 National Oceanic and Atmospheric Administration. Table V compares the data from these sources. From Table V, the general conclusions can be drawn that 1) 1973 open ocean surface temperatures were lower than average for that area. 2) 1973 Fort Point sea surface temperatures were higher than the Fort Point average 3) In 1973, the open ocean surface temperatures were lower than those at Fort Point, except during December-March. It was decided not to utilize the Fort Point data since that station is situated at the southern base of the Golden Gate bridge. It was felt that those measurements were not representative of more exposed, deep water locations, where upwelling^waters interact to form fog. / The mean monthly measurements for 1973 (column (1)) were chosen because: 1) Fishing Information is a publication readily available to anyone requiring such data, a <' 2) The isotherm charts allow the selection of a representative temperature for the ocean surface over which the prevailing north- westerly winds actually blew, and 4 3) These data represent the period of study and are not modified by data from other years as in long term averages. 34 cu -d +» u o «+-l 00 -UJ c S* w * £3 a. "5 S (U *H H H o «> 3 H * 42 a) d cu o ^ ^ ^ in ■t-i u « O co o > CO On CO O H Oirii— "cxDooor^oocovor^-m OOOO^HCOCVJCOCNJ^O'-* O^OOO"— 'OOOOOO + + + I H- + + + + + + co n co -h (\j oovor^-co— «co HfMNNfnininmui't(«ii-i TfOvo^Hc^t^rgiricooN o o h M oci^ininTjtflH C S 2 O £ d M O H en m^coc^as>— • r~ o o^ o o r— O O O O O <-h <— •CO'— i (\J n o i 4- + I l I I u* l I I i inflHOroHOcooorj ^in cMfNjrxjfMcMcO'H/iTfinm'^co CNJCNJCM-— • •— i N (M h co rn N N u (Tj a >> 0) cu c 3 ° 3 5 M CU 42 O o Z u cu 42 a ^ 5- a cu h H 5 <+H <+1 „ / Height of the inversion base (BI) Fog days for Monterey Airport, Point Pinos, Pigeon Point, Pillar Point Temperature at the top of the inversion (T ) Moisture index (MI) Temperature index (TI) Hidden Hills maximum daily temperature Hidden Hills minimum daily relative humidity Upwelling Index (UI) 36 The height of the inversion base has been noted by numerous authors as being of prime importance in fog formation. In Leipper's 1948 paper, it was listed as one of three non-diurnal indices which characterized fog, and one which could be used to predict fog on a day- by-day basis. On the graph depicting the height of the inversion base (Figures A- 1 and A-2), a tick mark along the horizontal axis indicates that there was no inversion at Oakland. The temperature at the top of the inversion (TT), based on the 0400 Oakland sounding, was used in calculating the temperature index, and varied directly with it. Tm by itself gives an indication of the strength of the inversion gradient because sea- surface temperature varies only by a few degrees. The moisture index (MI) wars cited by Leipper [1948] as a 4* critical index in describing fog development. The moisture index is the 1600 surface dewpoint temperature at Oakland minus the sea -surface temperature. The sea-surface temperature used in this calculation was described in the previous section. The time of 1600 was chosen as the time closest to the maximum on-shore flow of marine air. One can think of a high positive moisture index as indicating a high degree of saturation in the surface air, and a condition favorable to fog formation. The temperature index (TI) is the third important descriptive index in fog development. The temperature index is the 0400 temperature at the top of the inversion minus the sea- surface temperature. TI is a 37 measure of the strength of the inversion gradient and of overall atmos- pheric stability. A high TI means relatively warm conditions aloft and a strong gradient, which is favorable to fog formation. Since upper-air observations are few, it was decided to use those taken at a nearby inland location. Hidden Hills provided an estab- lished, convenient location for recording 24-hour continuous temperature and relative humidity at a substantial elevation above sea level. Hidden Hills is located about five miles inland from Monterey Bay, at a height of 262 m. By monitoring the parameters of maximum daily temperature and minimum daily relative humidity, it was possible to determine whether a continental regime or the marine layer was enveloping Hidden Hills. High temperatures and low relative humidities were indicative of warm, dry continental air", while a reversal of the para- meters indicated cool, moist marine air at this intermediate altitude. Bakun's [1974] upwelling index (UI) was calculated using six-hourly synoptic surface atmospheric pressure fields. The greater the northerly wind component, the greater wa^'the degree of upwelling, and the cooler the sea-surface temperature; hence the more favorable were conditions for fog. An index value greater than zero indicates upwelling, while less than zero signifies downwelling^ In cases of only slight upwelling, there may have been no effect on sea- surface temp- erature if the mixed layer were of moderate depth. 38 2. Daily Data Appendix B, 1973 Daily Data includes the following: Oakland radiosonde - 0400 (local) Continuous surface visibilities for Monterey Airport, Point Pinos, Pigeon Point, and Pillar Point. These data are presented with the Oakland soundings across the top of each page. The visibilities at the coastal stations for the same dates are aligned beneath each sounding plot. The visibilities are plotted in a bar graph format, allowing comparison on an hourly basis and a direct comparison of one day to other days. Each bar graph represents a 24-hour period (1600-1559 PST) for each station. Each 0400 sounding can be correlated with the daily bar graphs directly, and the visibilities at any time can be compared between stations. The visibility data for the Monterey Airport were recorded as conditions changed, often several times an hour. The Monterey Airport data were considered highly reliable^/ The data from the three coastal stations consisted of observations entered every three hours. Using persistence, the conditions between observations were assumed to be the same as at the observation times. For cas'es in which the conditions were different for subsequent observations, the period between observations was divided in half and the appropriate visibility conditions assigned to each period. 39 The visibility graphs show four ranges of surface visibility: heavy fog, 0 - visibility - 1/2 mile light fog, 1/2 < visibility - 3 miles |::. *:: | haze, 3 < visibility - 7 miles I | clear, visibility 7 miles A vertical line ( [ ] j ) was used together with one of the above symbols to indicate any occasion when overcast was reported on the surface weather observation sheets. Although visibility less than one kilometer is the interna- tionally accepted definition for fog, visibility of one-half mile was used as the cut-off criterion in this paper because all stations reported visibility in units of miles vice kilometers. Visibility data are presented for all fog sequences which occurred in 1973, and the 0400 Oakland radiosonde is plotted for every day of the year. 40 V. DATA ANALYSIS A. YEARLY ANALYSIS From figures A- 1 and A-2, it is seen that 1973 can be divided into two fog- related periods. The summer fog period is taken to be late April through October and the winter fog season November through mid-April. The fog- related periods are designated based on the number and distribution of fog days throughout the year, as shown in Figures A-l and A-2 and in Table VI. As used in this analysis, the term "fog" and "fog-day" refer to visibility conditions less than one-half mile. When the term "light fog" is used, it refers to visibility conditions greater than one-half mile and equal to or less than three< miles. TABLE VI 1973 Marine Fog Occurrence at Central California Coastal Stations Summer fog days* Winter fog days Summer fog hour's Winter 'fog hours Summer average daily duratioc(hr) Winter average daily duration(hr) Monterey Airport 47 9 207 29 4.0 3.7 Point Pinos 62 11 485 73 7.8 6.6 Pigeon Points- 48 4 265 21.* 5.5 5.2 Pillar Point*** 91 17 784 167 8.6 9.8 * Days on which fog was reported ** Missing 20 days of data in summer and 52 days in winter *** Missing 4 days of data in summer and 58 days in winter. 41 To check on the validity of the 1973 data, the percentage of hours with fog for the stations at Point Pinos, Pigeon Point, and Pillar Point were calculated and compared with mean data presented by Patton [1956] based on figures supplied by the Coast Guard. (Pillar Point was not listed by Patton; however, since Point Montara is in the immediate vicinity, Pillar Point was compared to Point Montara). Table VII compares the results. TABLE VII Percentage of Hours with Fog at Coastal Stations 1973 Mean year (USCG data) Point Pinos 6.4 9.0 Pigeon Point 3. 3 11.8 Pillar Point 10. 9* 10.6 (Point Montara) It should be recalled that the 1973 data were not complete; most notably, Pigeon Point was missing almost 20 days in the summer fog period. However, even assuming that Pigeon Point experienced the average daily fog duration of 5. 5 hours for the 20 days for which data are missing, this would raise the yearly total hours to 395, or only 4.5%. This figure appears unreasonably low, especially with the higher figures for Point Pinos and Pillar Point. These observations serve to reinforce the author's opinion that the surface observations for Pigeon Point are incomplete and probably inaccurate. 42 Based on the location of Pigeon Point and on the daily record of visibility at the two adjacent stations, it is estimated that Pigeon Point probably experienced an additional 25 days of fog, that is, 14 in the summer period and 11 in the winter. It is impossible to estimate accurately the additional fog in hours that may have occurred at Pigeon Point. However, by using a daily average duration of 8. 0 hours for the summer and 7. 5 hours for the winter and multiplying by the estimated number of additional days, a rough figure can be determined (the values of 8. 0 and 7. 5 hours were obtained by choosing a representative figure intermediate between the durations of Point Pinos and Pillar Point). The above estimations and computations yield the revised figures for Pigeon Point: 1 Summer fog days Winter fog days. Summer fog hours Winter fog hours Summer average daily duration(hr) Winter average daily duration (hr) Pigeon Point 62 15 377 104 6.0 6.9 This compares more favorably with the two adjacent stations, although the values given for the summer remain somewhat low. It is seen from Table VI that the summer period had many more days on which fog occurred and many more total hours of fog at all stations than did the winter period. Additionally, the data indicate a generally longer average daily duration of fog during the summer. Using this adjusted Pigeon Point data, it may be shown that the average 43 daily duration of fog at each station increases with distance north except at Pigeon Point, in the summer period. The greatly decreased daily duration of fog at Monterey Airport is undoubtedly due to the diurnal heating of the land causing accelerated dissipation of the fog. This illustrates the difference which may occur between a station with a complex environment only 1. 5 miles from the water, and a true coastal station. Several compilations of parameters from Figures A- 1 and A_2 have been made to show the difference between summer and winter, and are presented in Tables VIII-XI. Table VIII compares inversion base (BI) statistics for the two fog-related seasons. TABLE -VIII / Inversion Base Statistics - Oakland, 1973 Summer Winter 400 m < BI < 1500 m 78 days (41. 0%) 20 days (11. 5%) 0 < BI ^ 400 m 39 days C20:'5%) 8 days (4. 6%) BI = 0 67 days (35.3%) 113 days (64. 9%) No inversion at Oakland 6 days (3.2%) 33 days (19.0%) Total days* 190 days 4 174 days *Total days = 364 due to one missing radiosonde 44 It must be recognized, however, that Oakland exhibits meteoro- logical features more characteristic of a land station than a coastal station. Because Oakland's location is inside the mountain range that separates most of San Francisco Bay from the ocean, the Oakland radiosonde is not indicative of the low level atmospheric conditions over the open ocean where advection fog is formed. It has therefore been assumed that if a no-inversion condition is shown by the 0400 Oakland sounding and the surface air temperature at Oakland (T ) aoak is greater than the ocean surface temperature (T ), an inversion must s exist over the open ocean due to the cooling of the air by the surface waters. If, however, T_ / T , then a "no-inversion" situation aoak ^ s exists over the ocean as well. The days of no inversion at Oakland are indicated on Figures A-l and A-2 on the graph entitled "Height of the Base of the Inversion" by a tick mark below the horizontal axis. These cases were adjusted, where necessary, to open ocean conditions and are shown in Table IX. In 1973 there were 39 days of no inversion!' at Oakland, but as seen from Table IX, 12 of these were cases in which T ^ T , s oak or an open ocean inversion was assumed to exist. There were seven reported fog days which occurred during the 39 days qrt no inversion at Oakland. Of these seven fog days, five occurred on days of assumed open ocean inversions. In other words, during the days of no-inversion at Oakland, 71% of the fog occurred in 31% of the time. The remaining 45 X! i— t W CQ < Eh £3 nJ f-t 44 «J 0 _ -^ ■4.) fH CO • 1-1 O OJ -? o *»■» -l-> o c o 0 «* •l-< o CU J-l — G 43 • 1-1 u o • I-l 43 G * X5 cu c O CO 43 rt d cd 43 •i-i '-H — < CO • iH p CJ cu 5 o .2 0 « c u oo Tf o .2 a h o c CU 4-> 00 x G CO *« M ?h oo ° m i-H .2 h CM o > Z o c o ffj rt ro oce sio. 9. CU CU CJ ? > o (VJ i— 1 in a u '> ■P' o <5 G O • |H CO Jh iD % G o O CD XI >> G O J4 O g o CO nJ cu CO 42 G O •l-l 4-1 42 •|H u 4-1 CO •H p d a) CU g 0 , I a CJ o • I-l CO o u • H — u 0 •i-l cu % x> > G o •i-i O CJ §,-' G •«H « 0 CQ CJ • H *H 0 CO CU h > G cu a •i-i ft |>s £3 0 o «J •^ a II T) 42 ii CO cd 0 0 b] ir> ctf <*-( • H 4-1 «j Eh 0 3 cu S/ (h 42 •i-i u 4J CQ CJ 0 u CU 0 V 1 • i-l > at CO 3 P o Eh H 2 w 46 two fog days, occurring during conditions of T ^. T , appear oak s to have been the result of frontal effects, since cold fronts were in the near vicinity of the San Francisco Bay area on both occasions [Weatherwise 1974]. Based on data which have been adjusted in Table IX, Table VIII can be revised to reflect the assumed conditions for the open ocean off the Golden Gate, as follows (Table X) TABLE X Inversion Statistics - Open Ocean off San Francisco 1973 Summer Winter 400 m< BI < 1500 m 78 days (41.0%) 20 days (11.5%) 0 < BI * 400 m 39 days (20.5%) 8 days (4. 6%) BI = 0 7fc days (37.5%) 121 days (69.5%) No inversion 2 days (1.0%) 25 days (14. 4%) Total days * 190 days 174 days * Total days = 364 due to one missing radiosonde During the winter, either a surface inversion (BI=0) or no inver- sion existed 83. 9% of the time (146 days) at Oakland. For only 4. 6% of the winter season (8 days) was the inversion base between 0 and 400 m. According to Leipper [1948] a surface inversion represents the phase of the fog sequence immediately prior to the actual formation of fog. 47 Once fog forms, the mixed layer should cool due to radiation and sub- sequent mixing and the inversion should lift off of the surface. Except for the eight days during which the inversion base was between the sur- face and 400 m, conditions in the winter season were generally unfavor- able for fog formation. In other words, _if fog formed at sea during the the winter fog period, it was shallow (because the inversion base rarely rose above the surface) and undoubtedly dissipated before being advected ashore. Conversely, the summer shows conditions much more evenly distributed. The important requirement, noted by Leipper [1948], that the inversion lift off of the surface in order that fog may be formed and advected, occurred almost five times as frequently in the summer as in the winter. Additionally, the no inver-aion condition, which is not con- ducive to fog formation, occurred only 1. 0% of the time in the summer against 14.4% in the winter. The overall picture, based on the height of the inversion base, is one of a far more favorable environment for advective fog formation at the coast in the summer than in the winter. Referring to Figures A-l and A-2, the mean temperature at the Q o top of the inversion (Trp) was 18. 8 C for the summer and 10. 2 C for the winter. Since the sea surface temperature (Ts) t^nds to equal the air temperature directly above it, and since the seasonal variation of T is more stable than the variation of T , then the higher TT indicates a stronger inversion gradient, which serves to prevent mixing of the marine layer with the upper air. 48 o o The mean moisture index was +0. 3 C for the summer and -3.3 C for the winter. This index, calculated daily at the radiosonde time of maximum onshore sea breeze, indicates the relative amount of moisture in the surface air layer offshore. Leipper [1948] cited a moisture index favorable to fog at San Diego in winter as any positive value or any o negative value between 0. 0 and -5.0 C. When the daily temperature indices were averaged, the mean temperature index was 7. 8 C for the summer and -1. 4 C for the winter.' The temperature index, calculated at the radiosonde time closest to the time of maximum off shore land breeze, points out the influence of the warm, dry continental upper air in strengthening the inversion. Leipper [1948] listed any positive ( + ) value of the temperature index as a favorable condition for marine fog a^San Diego. A summary of the parameters listed above is presented in Table XI. Table XI supports the contention that one must observe trends, rather than absolute values when dealing with fog formation. The mean temperature at the top of the inversion was 18. '8 C for the entire sum- o - o mer, and 19.5 C for the summer days with fog; only 0. 7 C difference. Yet, as seen in Figures A-l and A-2, large and definite upward and downward trends are obvious. The same may be saidf for the other indices. With coastal upwelling linked so closely to marine fog formation, it would be expected that maximum upwelling would occur during the 49 X! W ffl < Eh CO o CO •i-i u > rtJ XJ CD sO c\J U 0 o d i— t U o CM CM i u o (M d i Winter (174 days) CO >» XJ CO CO XI i— i U 0 CNJ © i—4 U o CO CO 1 U 0 <* i Summer fog days (111 days) CO XI CO CO xt 1— 1 1— 1 U o IT) u O 00 p U o 00 00 Summer (190 days) CO nJ XJ O^ CO -V — » CO (d XI o o^ 1—1 U o CO 00 r— 1 u o CO o u o 00 5> T3 ti (0 JA a O -t-> a o o SI cq V o Mean temperature at top of inversion* 0) XI a cu CD •l-l o a d a>