EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 1

eint vol7_0008 Mp_1 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 1

Copyright q 2003. Paper 7-008; 4,059 Words, 7 Figures. http://EarthInteractions.org

The Santa Ana Winds of California M. N. Raphael Department of Geography, University of California, Los Angeles, Los Angeles, California Received 21 January 2003; accepted 17 March 2003

ABSTRACT: A 33-yr, numerical dataset of the occurrence of Santa Ana winds for the period 1968–2000 has been created and validated. Daily Weather Maps were examined to identify the days when a surface high pressure system existed over the Great Basin simultaneously with a surface low pressure system offshore of southern California, and the prevailing wind over southern California was from the northeast quadrant. The dates of these occurrences, as well as the wind speed, temperature, and dewpoint temperature among other variables, were extracted and tabulated. The frequency of Santa Ana events derived from the weather maps was compared to events defined by wind direction only and there is agreement between the two. Preliminary results show that the Santa Ana event is limited to the period September–April and that the month of peak occurrence is December. The average frequency of events is 20 yr21 and the average duration of an event is 1.5 days. Humidity levels are not uniform across Santa Ana events; the driest months are the months with the highest frequency of events. The frequency of Santa Ana events is usually lower than average during El Nin˜o events. These preliminary results indicate that the dataset is useful for in depth study of the local phenomenon and its effect on the region within the context of the large-scale circulation. KEYWORDS: Santa Ana winds * Corresponding author address: Marilyn Raphael, Department of Geography, University of California, Los Angeles, 1255 Bunche Hall, Los Angeles, CA 90095-1524. E-mail address: [email protected]

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 2

eint vol7_0008 Mp_2 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 2

1. Introduction The Santa Ana wind is a hot, dry, foehn-type, easterly or northeasterly wind that blows from the deserts east of the Sierra Nevada to the coast of southern California (Glickman, 2000). It tends to occur in winter and spring. While it is named after the pass and river valley of Santa Ana, California, it can affect much of the southern California region. The occurrence of the Santa Ana wind is anticipated each year. It is an important local, meteorological phenomenon commanding scientific (e.g., Mensing et al.,1999) and social (e.g., Miller, 1968) interest because of its relationship to forest fires (e.g., Minnich, 1983; Keeley and Fotheringham, 2001), and therefore to watershed runoff, its effect on temperature, humidity, and on the distribution and deposition of air pollutants. The Santa Ana wind has been the subject of study; however, much of that work is quite dated (as the reference list indicates) and is in the form of case studies of the local characteristics of individual events (e.g., Fosberg, 1965). Its three-dimensional physical structure has been studied (Fosberg et al., 1966) and the windflow pattern over its region of influence has been mapped (e.g., Edinger et al., 1964). These studies neither attempt to evaluate the temporal variability of the winds nor to assess the role of the larger-scale circulation. There has not been any recent attempt to assess change. The Santa Ana wind has been defined in terms (individually or combined) of its speed and direction, local pressure gradients, associated temperature, and relative humidity. For example, Edinger et al. (Edinger et al., 1964) for their 7yr dataset based their definition of a Santa Ana event on the occurrence of a 3mb drop in pressure from Palmdale to Santa Monica, California; a 98C or greater increase in temperature from Santa Monica to Palmdale; northerly winds with speeds of 30 mph or greater in the Riverside, San Benardino Valley, California; and a relative humidity of 30% or less in the Los Angeles, California, area. If one or all of these criteria were met, then the data were subjected to further analysis. Sergius and Huntoon (Sergius and Huntoon, 1956) used relative humidity, wind speed, and direction to define the Santa Ana wind. If the wind speed averaged over four intervals during the day was greater than or equal to 20 mph, the wind direction was from the northeast quadrant, and the relative humidity at 1630 PST (Pacific standard time) was less than 40%, it was considered to be a Santa Ana event. Richardson (Richardson, 1973) based his definition on the pressure gradient between Las Vegas, Nevada, and San Diego, California. The variables used in these definitions are important locally, and while useful for local characterization of the wind, they do not allow an analysis of the Santa Ana wind’s temporal variability within the context of its dynamic causes. The dataset presented here will allow interpretation of the winds within the larger, regional and global-scale climate. Until now, there has been no consistent long-term record, and the record that exists does not define the Santa Ana winds in terms of the large-scale circulation. This is unfortunate because the large-scale pressure distribution provides the necessary conditions that initiate the Santa Ana winds (Sergius et al., 1962). The necessary conditions that give rise to the Santa Ana winds are the existence of an anticyclone in the Great Basin high (GBH) simultaneously with a surface low

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 3

eint vol7_0008 Mp_3 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 3

pressure system off the California coast. Schroeder et al. (Schroeder et al., 1964) showed that the GBH develops when a cold front associated with a trough aloft passes through California. As high pressure builds over the Great Basin a strong northeast-to-southwest surface pressure gradient is established over southern California (Sommers, 1978). Typically, the GBH has a cold core so most of the Santa Ana winds are of the cold type (von Ficker and Rudder, 1943). However, during early fall, the warm type is sometimes observed. The synoptic conditions described above are peculiar to the period from fall to early spring when the surface high pressure, which dominates the southwest coast of the United States from spring to summer, moves farther south, as is typical of the Mediterranean climate regime. Also in summer, the monsoonal flow over the southwestern United States, coupled with the intensification and northward movement of the Pacific high, promotes westerly flow at local and regional levels. Therefore, Santa Ana events rarely, if ever, occur in summer. The dataset reported on here characterizes the frequency of occurrence of the Santa Ana winds based upon the large-scale regional, atmospheric dynamics that initiate the winds. It contains detailed information on wind speed, temperature, pressure, and humidity. It was created primarily to facilitate the investigation of the regional, dynamic controls on the development, movement, and duration of the high and low pressure systems into the positions that allow the initiation of the Santa Ana winds. The creation of the dataset is described in section 2. Some initial results and possible avenues of research are outlined in section 3, and a summary is presented in section 4.

2. Dataset description The fundamental assumption underlying the creation of the dataset is that the occurrence of the Santa Ana winds depends upon the presence of a high pressure system over the Great Basin simultaneously with a low pressure system offshore of southern California (e.g., Figure 1). Additionally, the local winds should blow predominantly from the northeast quadrant. Accordingly, daily synoptic weather maps obtained from the National Climate Data Center for the years 1968–2000 were used to select days when the preceding conditions prevailed. These daily maps represent the surface weather at 0400 PST. Fortuitously, the Santa Ana wind is best developed at night or early morning when the influence of the sea breeze is gone and the land breeze augments the flow (Fosberg et al., 1966). The land/ sea breeze is an important component of the surface wind regime in southern California. By day the sea breeze penetrates inland moderating temperature, humidity, and pollution levels. By night the land breeze develops with consequent return flow to the coast. During Santa Ana events the sea breeze blows in the opposing direction to the northeasterly wind. The Santa Ana wind speed varies but tends to be strongest in the early hours of the day before the sea breeze is well developed. The net result is a reduction in the speed of both winds and reduced inland penetration by the sea breeze in the late afternoon (Fosberg et al., 1966). Each map was visually scrutinized to determine if the required surface pressure distribution and wind direction existed. Central surface pressure difference

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 4

eint vol7_0008 Mp_4 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 4

Figure 1. Example of large-scale pressure distribution used in defining a Santa Ana day.

between the GBH and the southern California low ranged from 4 to 28 hPa. If the GBH and the surface low did not exist concurrently, which can happen even during the cooler months, it was not considered to be a Santa Ana day. These restrictions resulted in the elimination of all days from May to August. This is consistent with what is expected from the surface pressure regime of the region as discussed in section 1. Local meteorological conditions at Los Angeles were used during the selection process. Los Angeles is a heavily urbanized region; therefore, the temperature associated with this dataset may be influenced by its urban heat island. Note, however, that the region has been urbanized over the lifetime of the dataset. The

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 5

eint vol7_0008 Mp_5 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 5

effect of urbanization is largely on the temperature, but surface heating plays only a minor role during Santa Ana events. There is little diurnal variation in temperature during Santa Ana events, and temperature change is dominated by descending air and associated adiabatic warming (Fosberg et al., 1966). Use of the Daily Weather Maps (1968–2000) means that the dataset has a qualitative basis. To counteract some of the imprecision that might accompany the process, each day that was chosen as a Santa Ana day was carefully assessed to ensure that the fundamental criteria were met. In addition, the imprecise boundaries of the high and low pressure systems pose potential problems when an objective method is used. These problems are avoided by being able to locate the systems visually. Once the criteria were met several variables were extracted including wind speed, wind direction, dates and number of days per event, dewpoint temperature, actual temperature, as well as central pressure in the high and low pressure systems and their general location and spatial extent (latitude and longitude). The number of days during which Santa Ana winds prevailed in the dataset described above was calculated and compared (on seasonal and monthly timescales) to the number of Santa Ana days defined by wind direction (from the northeast quadrant) at a meteorological station located in Tustin, California (see Figure 1). Tustin lies east of the Santa Ana Mountains (after which the winds were named) and is subject to very strong winds when Santa Ana events occur. The station data were obtained from EarthInfo, Inc. (EarthInfo, Inc., 1995). Figure 2 shows the frequency time series of Tustin and the new dataset for the period 1968–95. In the weather map–derived data there are fewer Santa Ana days, and the number of days appears more variable than the Tustin station data. The differences in the number of days are expected since it is possible to have a wind from the northeast quadrant in the absence of the large-scale pressure distribution required by the map dataset. The local land/sea-breeze circulation has a significant impact on low-level wind direction at night when the offshore wind is well developed (Fosberg et al., 1966). However, in general, there is good agreement in the timing of the occurrence of events. For example, they both show decreasing numbers of event days over the period 1970/71–1979/80. Although Figure 2 suggests that the two time series generally vary together, they are not well correlated. The monthly analysis shows that the weak correlation arises in the months October, December, and February. In the other months there is moderate to strong correlation with r 5 0.5–0.7.

3. Summary of Santa Ana statistics 3.1 Event frequency In this dataset Santa Ana events are seasonal; that is, they begin in September and end in April (Figure 3). The month of highest frequency of event is December, closely followed by November. The pattern of occurrence is asymmetric, indicating that late fall/early winter is the favored time of occurrence of Santa Ana winds. This general monthly pattern of frequency has also been found in studies based on wind speed and direction alone (Sergius and Huntoon, 1956; Edinger

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 6

eint vol7_0008 Mp_6 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 6

Figure 2. Number of days during which Santa Ana events were recorded based on wind direction at Tustin vs number of days compiled using the daily synoptic weather maps.

et al., 1964; Richardson, 1973). This is the time of year when the passage of low pressure systems over southern California is most frequent. Since these lows and their embedded cold fronts are part of the synoptic conditions that allow Santa Anas to develop (Schroeder et al., 1964), the monthly frequency of events is reasonable. The dates of onset and cessation of the events are part of the present dataset so that interannual and longer-term variations of the Santa Ana season may be explored. The fact that this climatology is consistent with anecdotal and published research indicates that the present dataset is a robust one and will prove useful for further study of the Santa Ana phenomenon. The frequency of Santa Ana wind events per season is shown in Figure 4. On average, 20 events occur each season with a standard deviation of 5. From 1973/74 to 1982/83 the frequency of events is consistently lower than average, suggesting that the large-scale pressure distributions during that period differed from those earlier and later in the dataset. The average duration of an event is 1.5 days (Figure 5), and initial analysis of the interannual variation of this variable suggests that in the early 1970s events appear to last longer than the early 1980s. Comparison of Figures 4 and 5 shows that during the period of low frequency of events, the duration of an event was shorter.

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 7

eint vol7_0008 Mp_7 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 7

Figure 3. Average number of occurrences of Santa Ana events per month.

3.2 Dewpoint temperature One important characteristic of the Santa Ana winds is that they have very low humidity levels. The average dewpoint temperature (DPT) per Santa Ana event per year and per month is used as a measure of humidity (Figure 6) . During the Santa Ana events in this dataset the annual-average DPT ranges from approximately 0.58 to 88C with a mean value of 3.68C. In the literature, relative humidity is the variable usually used to describe humidity levels during a Santa Ana event and is not directly comparable to the DPT. However, Figure 6a shows that the humidity levels associated with the Santa Ana events are not uniform and appear to have interannual and longer period variations that may be explored. The average monthly dewpoint temperature (Figure 6b) shows that there is a continuous decrease in humidity from September to December, followed by a continuous increase in April. The driest months are December and January and the most humid by far, is September. Together, Figures 3 and 6b show that the driest months are those during which Santa Ana events are most frequent and the most humid months are those when the Santa Ana winds are least frequent. This is interesting and previously unknown information and may be related to seasonally changing moisture conditions in the GBH as well as over the southern California region. For example, if the moisture content of the air is a function of the moisture evaporated by the descending air as it moved toward and over southern

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 8

eint vol7_0008 Mp_8 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 8

Figure 4. Frequency of Santa Ana events per season per year. The ‘‘WE’’ indicates an El Nin ˜ o winter.

California, then the monthly variation in humidity levels indicates in part the preceding summer dryness of the surface followed by the increasing wetness in the winter and spring months. It may prove useful to compare the humidity levels during Santa Ana events with that of the climatological monthly averages. 3.3 Santa Ana and ENSO Initial spectral analysis of the frequency of the events shows a significant spectral peak at approximately 3.8 yr (Figure 7). The occurrence of ENSO events vary within this same timescale so it is of interest to see if there is some relationship between Santa Ana events in this dataset and ENSO. In Figure 4, a ‘‘WE’’ indicating a warm event identifies the season during which an El Nin˜o event, in its mature stage, occurred. Most of the warm events are associated with belowaverage frequency of Santa Ana events. To explore this further, the Southern Oscillation index (SOI) from September to April was correlated with Santa Ana event frequency at both monthly and seasonal levels of aggregation. At the seasonal timescale the time series are not well correlated; however, there is significant positive correlation in February and March. This suggests that if Santa Ana events

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 9

eint vol7_0008 Mp_9 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 9

Figure 5. Average duration of Santa Ana events each year.

are modulated by ENSO, the influence is obvious only late in the season. Also, the positive relationship supports the information given in Figure 4. It indicates that during El Nin˜o seasons (when the SOI is negative) the tendency is to have fewer Santa Ana events. Intriguingly, El Nin˜o events appear associated with longer-lasting Santa Ana events. Compare the information in Figure 4 with the time series of weather map–derived data in Figure 2. Bearing in mind that ENSO is not a stationary system, that is, there are substantial differences between events (e.g., Diaz and Kiladis, 1992), during an El Nin˜o winter there is the tendency for more cyclones (with their attendant precipitation) to move into the southern California region. This is associated with the expansion of the Aleutian low toward the west coast of the United States (Schonher and Nicholson, 1989; Raphael and Mills, 1996). Given this information alone, more frequent Santa Ana events might be expected during an El Nin˜o winter given the synoptic conditions that precede the development of an event. At least initially, this dataset suggests otherwise. Further analysis should provide reasons for this apparent relationship. The foregoing represents only the results of preliminary exploration of the dataset. There is much more to examine. In addition to the variables described above there are those mentioned in section 2. These data can be retrieved at annual, monthly, or daily time steps thereby allowing extensive temporal vari-

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 10

eint vol7_0008 Mp_10 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 10

Figure 6. Dewpoint temperature: (a) average each year and (b) average by month. Units: 8C.

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 11

eint vol7_0008 Mp_11 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 11

Figure 7. Spectrum of Santa Ana event frequency.

ability analyses. The criteria upon which the dataset compilation is based allows Santa Ana events to be integrated into the broader climatological picture of the West Coast climate. For example, the Santa Anas occur during the time of year when southern California receives most of its precipitation as well as when ENSO events affect precipitation in the region. The information in this dataset combined with information on the precipitation variability of the region would allow an analysis of the interaction between Santa Ana event frequency and precipitation. Additionally, any in-depth frequency analysis of the Santa Ana events in this dataset is necessarily a frequency analysis of the large-scale pressure distribution and therefore allows a link between them and atmospheric patterns that affect the entire region.

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 12

eint vol7_0008 Mp_12 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 12

4. Summary This paper describes a new dataset of the Santa Ana winds of California. It was compiled from daily synoptic weather maps using the criteria of simultaneously existing high pressure over the Great Basin, low pressure off the coast of southern California, and wind direction predominantly from the northeast quadrant. The nature of the information presented here is preliminary, but the dataset appears to be useful for frequency analyses as well as the study of dynamic aspects of the Santa Ana winds. So far the results establish that the Santa Ana event is limited to the period September–April and the month of peak occurrence is December. The driest (most moist) months are the months during which Santa Ana events are most frequent (least frequent). There is substantial interannual variation in frequency and duration of these events and humidity levels also exhibit interesting interannual and longer-term variability. There appears to be a relationship between ENSO events and Santa Ana event frequency such that fewer events occur during El Nin˜o seasons. This relationship appears complex and demands further analysis. This information is new and is only the beginning of what might be gleaned from this dataset. There is the potential for examining the interaction between Santa Ana events and larger-scale atmospheric phenomena that influence the climate of the region. Acknowledgments. This research was funded in part by the UCLA Academic Senate and the UCLA Institute of the Environment. I thank Leonard Tang, Larry Chang, Sigrid Rian, and Aaron Potito for their able assistance with compiling and analyzing the dataset. I also thank Chase Langford for his assistance with the diagrams and two anonymous reviewers for their helpful criticisms.

References Daily Weather Maps, 1968–2000. NCDC Cooperative Station Data, NCDC, CD-ROM. Diaz, H. F., and G. N. Kiladis, (Eds.), 1992: Atmospheric teleconnection patterns associated with the extreme phases of the Southern Oscillation, in El Nino: Historical and Paleoclimatic Aspects of the Southern Oscillation, Cambridge University Press, New York, New York, 7– 28. EarthInfo., Inc., 1995: COADS Global Marine. NCAR, CD-ROM. Edinger, J. G., A. R. Helvey, and D. Baumhefner, 1964: Surface wind patterns in the Los Angeles Basin during Santa Ana conditions. Department of Meteorology, UCLA, Part I of Final Report on Research Project 2606, Suppl. 49, USFS-UC Contract A5fs-16563, 71 pp. Fosberg, M. A., 1965: A case study of the Santa Ana winds in the San Gabriel Mountains. U.S. Forest Service Research Note PSW-78, Pacific Southwest Forest and Range Experiment Station, Berkeley, CA, 11 pp. Fosberg, M. A., C. A. O’Dell, and M. J. Schroeder, 1966: Some characteristics of the threedimensional structure of Santa Ana winds. U.S. Forest Service Research Paper PSW-30, Pacific Southwest Forest and Range Experiment Station, Berkeley, CA, 26 pp. Glickman, T., 2000: Glossary of Meteorology. 2d ed. Amer. Meteor. Soc., 855 pp. Keeley, J. A., and C. J. Fotheringham, 2001: Historic fire regime in southern California shrublands. Conserv. Biol., 15, 1536–1548. Mensing, S. A., J. Michaelson, and A. Byrne, 1999: A 560-year record of Santa Ana fires reconstructed from charcoal deposited in the Santa Barbara Basin. Quat. Res., 51, 295–305.

EARTH INTERACTIONS (AMS) Friday Jul 11 2003 08:51 AM Allen Press • DTPro System

Earth Interactions

GALLEY 13

eint vol7_0008 Mp_13 File # 01TQ

• Volume 7 (2003) • Paper No. 8 • Page 13

Miller, W. H., 1968: Santa Ana winds and crime. Prof. Geogr., 20, 23–27. Minnich, R. A., 1983: Fire mosaics in southern California and northern Baja California. Science, 219, 1287–1294. Raphael, M. N., and G. M. Mills, 1996: The role of mid-latitude cyclones in the winter precipitation of California. Phys. Geogr., 48, 251–262. Richardson, R. T., 1973: The continental contribution to the climate of southern California. Ph.D. dissertation, Department of Geography, University of Oregon, 195 pp. Schonher, T., and S. E. Nicholson, 1989: The relationship between California rainfall and ENSO events. J. Clim., 2, 1258–1269. Schroeder, M. J., et al., 1964: Synoptic weather types associated with critical fire weather, in Pacific Southwest Forest and Range Experiment Station, Berkeley Press, Berkeley, California, 264–274. Sergius, L. A., and J. K. Huntoon, 1956: An objective method for forecasting the Santa Ana. Forecaster’s handbook, U.S. Fleet Weather Central, San Diego, CA. Sergius, L. A., G. R. Ellis, and R. W. Ogden, 1962: The Santa Ana winds of California. Weatherwise, 15, 102–105. Sommers, W. T., 1978: LFM Forecast variables related to Santa Ana Wind occurrences. Mon. Weather Rev., 106, 1307–1316. von Ficker, H., and B. Rudder, 1943: Fohn und Fohnwirkungen. Der gegenwartige. Stand der Frage, in Probleme der Biometeorologie, edited by Becker and Erler, Leipzig, Germany, 112 pp. Earth Interactions is published jointly by the American Meteorological Society, the American Geophysical Union, and the Association of American Geographers. Permission to use figures, tables, and brief excerpts from this journal in scientific and education works is hereby granted provided that the source is acknowledged. Any use of material in this journal that is determined to be ‘‘fair use’’ under Section 107 or that satisfies the conditions specified in Section 108 of the U.S. Copyright Law (17 USC, as revised by P.L. 94-553) does not require the publishers’ permission. For permission for any other form of copying, contact one of the copublishing societies.