CalClim: California Climate Data Archive

CALIFORNIA CLIMATE WATCH
CURRENT ISSUE: February 2005

Read the full newsletter here: Word Format or PDF format

Spin, Spin, Spin

By Laura Edwards
Western Regional Climate Center

Tornadoes in the United States are most often associated with "Tornado Alley" through the central United States. The recent tornadoes in the Sacramento area on February 21st were a good reminder of tornado occurrence in the Central Valley. Although the Central Valley is a known region of tornado activity in California, they can really occur most anywhere, including the South Coast, Southeast Desert and Central and North Coast regions. What some may not know is that tornado frequency per unit area per time in the Los Angeles Basin is higher than the state of Oklahoma (1). In the Los Angeles Basin, for the period of 1950-1992, there were 3.19 tornadoes per square kilometer per year. In the state of Oklahoma, there were 2.86 tornadoes per square kilometer per year for the same period.

Figure 1. February 21, 2005 tornadoes near Sacramento, CA. From sacbee.com, photo by Matthew Adams.

This statistic may be surprising, but the qualities of the tornadoes that occur at each geographic area are different. First, most tornadoes in California, with the exception of the Southeast Desert region, occur in the cool season (winter). In the Plains, most occur in the warm season (spring-summer). Blier and Batten (1) found records of 242 tornadoes during the period 1950-1992 in California. Of these 242 tornadoes, 194, or 80.2%, occurred during the months of November to April. March had more tornadoes than any other month during their period of study with 55, or 22.7%.

Second, the pathlength and path width of tornadoes are shorter in California than central U.S. on average. Blier and Batten found an average pathlength of 1.5 miles and average path width of 98.0 yards. Other researchers have used the median values to describe tornado characteristics. Smith and Mirabella (2) found a median pathlength of 0.62 miles and path width of 43 yards for tornadoes in the period 1950-1971. For the same period in Iowa, the median pathlength was 4 miles and 170 yards.

Third, the average intensity of tornadoes is greater in the central U.S. than in California using the Fujita scale. The Fujita scale ranks tornado intensity by the amount of damage they cause, from F0 to F5 (see 3 for more information). Of the 242 tornadoes studied by Blier and Batten, 144 had F-scale values. 70 of these 144 were of F0 intensity, 50 were F1 intensity, 23 were F2 and one F3. This makes about 16.7% of these to be greater than F1 intensity. (Note: Currently, Storm Data at the National Climatic Data Center website has record of two F3 intensity tornadoes, one in Riverside county on August 16, 1973 and another in Orange county on February 9, 1978.) For the period 1950-1976, 38.3% of tornadoes in the contiguous United States were of F2 or greater intensity (4), whereas for the same period in California 26.7% of tornadoes were of F1 or greater intensity.

All of these statistics depend, however, on trained observers reporting the events. As a result, there are large discrepancies between various data sources. This is similar to the old riddle that asks if a tree falls in the woods, does it make a sound? If there is no one to report a tornado, then does it occur? The Australian Bureau of Meteorology has trained 5000 observers since the mid-1980s, and has found a significant increase in the number of tornadoes reported (5). In the decade from 1987-1996, tornado reports were a quarter of the total previously reported. The U.S. National Weather Service has a cooperative observer network, as well as regular training for storm spotters through many forecast offices. This program has helped ensure that trained observers are in as many locations as possible and to get the most accurate reports.

Figure 2 (left). Tornado on February 21, 2005 near Sacramento International Airport. From sacbee.com, photo by Stuart Hooper.

Hanstrum et al. (5) have also compared cool-season tornadoes in California with those of southern Australia. They showed that there are several similar meteorological aspects which could be used in better prediction of tornadoes in these regions. In the two cases, "the typical synoptic (10-100 km scale) pattern shows an active midlatitude trough just upstream, with a strong jet streak aloft. In both areas the tornadic thunderstorms occur with weak to moderate levels of thermodynamic instability in the lower atmosphere..." Another significant factor for F1-F3 tornadoes was large values of shear at low levels, within a kilometer or two of the surface. Shear is simply described as air moving faster aloft (i.e. one kilometer high) than at the surface, causing instability and turning of the atmospheric layer. These criteria and others were used in the winter of 1996 in Australia. This proved to successfully predict all nine tornado reports with the 24-hr forecast, and is now an operational forecast product. One primary different between southern Australia and California, however, is the topographic forcing (or lack of). The mountains of California's Coast Range and Sierra Nevada form a unique situation, but the "ingredients" listed above make a decent recipe for tornado forecasts.

This fascinating topic is always evolving and we are learning more with each California tornado, in addition to adding to our photo collections.

References:
1. Blier, W. and K. A. Batten, 1994. On the incidence of tornadoes in California. Weather and Forecasting, 9, 301-315.

2. Smith, T. B. and V. A. Mirabella, 1972. Characteristics of California tornadoes. Report to the University of California Lawrence Livermore Laboratory by Meteorology Research, Inc., 25 pp.

3. The Tornado Project website (Fujita scale page): http://www.tornadoproject.com/fscale/fscale.htm

4. Kelly, D. T., J. T. Shaefer, R. P. McNulty, C. A. Doswell III, 1978. An augmented tornado climatology. Monthly Weather Review, 106, 1172-1183.

5. Hanstrum, B. N., G. A. Mills, A. Watson, J. P. Monteverdi, C. A. Doswell III, 2002. The cool-season tornadoes of California and southern Australia. Weather and Forecasting, 17, 705-722.

For more information:
Sacramento Bee story on Feb 21, 2005 tornadoes: http://www.sacbee.com/content/news/weather/story/12425264p-13281620c.html

Roger Edwards of Storm Prediction Center, The Online Tornado FAQ: http://www.spc.noaa.gov/faq/tornado/

Jan Null, Golden Gate Weather Services Tornado Page: http://ggweather.com/climate/tornado.htm

Page last updated 2/11/05.

HOME	Data & Products	Monitoring & Forecasts	Related Research	Links	Site Map
Contact Us/Data Request Form	© 2004-2008 WESTERN REGIONAL CLIMATE CENTER. ALL RIGHTS RESERVED.

Climate Watch Archive All data in the Climate Watch is provisional and subject to change. Jan 2005 [PDF] [Word] Feb 2005 [PDF] [Word] For 2004 stories, click here Monthly Summaries by Bill Mork 2005 2004	CALIFORNIA CLIMATE WATCH CURRENT ISSUE: February 2005 Read the full newsletter here: Word Format or PDF format Spin, Spin, Spin By Laura Edwards Western Regional Climate Center Tornadoes in the United States are most often associated with "Tornado Alley" through the central United States. The recent tornadoes in the Sacramento area on February 21st were a good reminder of tornado occurrence in the Central Valley. Although the Central Valley is a known region of tornado activity in California, they can really occur most anywhere, including the South Coast, Southeast Desert and Central and North Coast regions. What some may not know is that tornado frequency per unit area per time in the Los Angeles Basin is higher than the state of Oklahoma (1). In the Los Angeles Basin, for the period of 1950-1992, there were 3.19 tornadoes per square kilometer per year. In the state of Oklahoma, there were 2.86 tornadoes per square kilometer per year for the same period. Figure 1. February 21, 2005 tornadoes near Sacramento, CA. From sacbee.com, photo by Matthew Adams. This statistic may be surprising, but the qualities of the tornadoes that occur at each geographic area are different. First, most tornadoes in California, with the exception of the Southeast Desert region, occur in the cool season (winter). In the Plains, most occur in the warm season (spring-summer). Blier and Batten (1) found records of 242 tornadoes during the period 1950-1992 in California. Of these 242 tornadoes, 194, or 80.2%, occurred during the months of November to April. March had more tornadoes than any other month during their period of study with 55, or 22.7%. Second, the pathlength and path width of tornadoes are shorter in California than central U.S. on average. Blier and Batten found an average pathlength of 1.5 miles and average path width of 98.0 yards. Other researchers have used the median values to describe tornado characteristics. Smith and Mirabella (2) found a median pathlength of 0.62 miles and path width of 43 yards for tornadoes in the period 1950-1971. For the same period in Iowa, the median pathlength was 4 miles and 170 yards. Third, the average intensity of tornadoes is greater in the central U.S. than in California using the Fujita scale. The Fujita scale ranks tornado intensity by the amount of damage they cause, from F0 to F5 (see 3 for more information). Of the 242 tornadoes studied by Blier and Batten, 144 had F-scale values. 70 of these 144 were of F0 intensity, 50 were F1 intensity, 23 were F2 and one F3. This makes about 16.7% of these to be greater than F1 intensity. (Note: Currently, Storm Data at the National Climatic Data Center website has record of two F3 intensity tornadoes, one in Riverside county on August 16, 1973 and another in Orange county on February 9, 1978.) For the period 1950-1976, 38.3% of tornadoes in the contiguous United States were of F2 or greater intensity (4), whereas for the same period in California 26.7% of tornadoes were of F1 or greater intensity. All of these statistics depend, however, on trained observers reporting the events. As a result, there are large discrepancies between various data sources. This is similar to the old riddle that asks if a tree falls in the woods, does it make a sound? If there is no one to report a tornado, then does it occur? The Australian Bureau of Meteorology has trained 5000 observers since the mid-1980s, and has found a significant increase in the number of tornadoes reported (5). In the decade from 1987-1996, tornado reports were a quarter of the total previously reported. The U.S. National Weather Service has a cooperative observer network, as well as regular training for storm spotters through many forecast offices. This program has helped ensure that trained observers are in as many locations as possible and to get the most accurate reports. Figure 2 (left). Tornado on February 21, 2005 near Sacramento International Airport. From sacbee.com, photo by Stuart Hooper. Hanstrum et al. (5) have also compared cool-season tornadoes in California with those of southern Australia. They showed that there are several similar meteorological aspects which could be used in better prediction of tornadoes in these regions. In the two cases, "the typical synoptic (10-100 km scale) pattern shows an active midlatitude trough just upstream, with a strong jet streak aloft. In both areas the tornadic thunderstorms occur with weak to moderate levels of thermodynamic instability in the lower atmosphere..." Another significant factor for F1-F3 tornadoes was large values of shear at low levels, within a kilometer or two of the surface. Shear is simply described as air moving faster aloft (i.e. one kilometer high) than at the surface, causing instability and turning of the atmospheric layer. These criteria and others were used in the winter of 1996 in Australia. This proved to successfully predict all nine tornado reports with the 24-hr forecast, and is now an operational forecast product. One primary different between southern Australia and California, however, is the topographic forcing (or lack of). The mountains of California's Coast Range and Sierra Nevada form a unique situation, but the "ingredients" listed above make a decent recipe for tornado forecasts. This fascinating topic is always evolving and we are learning more with each California tornado, in addition to adding to our photo collections. References: 1. Blier, W. and K. A. Batten, 1994. On the incidence of tornadoes in California. Weather and Forecasting, 9, 301-315. 2. Smith, T. B. and V. A. Mirabella, 1972. Characteristics of California tornadoes. Report to the University of California Lawrence Livermore Laboratory by Meteorology Research, Inc., 25 pp. 3. The Tornado Project website (Fujita scale page): http://www.tornadoproject.com/fscale/fscale.htm 4. Kelly, D. T., J. T. Shaefer, R. P. McNulty, C. A. Doswell III, 1978. An augmented tornado climatology. Monthly Weather Review, 106, 1172-1183. 5. Hanstrum, B. N., G. A. Mills, A. Watson, J. P. Monteverdi, C. A. Doswell III, 2002. The cool-season tornadoes of California and southern Australia. Weather and Forecasting, 17, 705-722. For more information: Sacramento Bee story on Feb 21, 2005 tornadoes: http://www.sacbee.com/content/news/weather/story/12425264p-13281620c.html Roger Edwards of Storm Prediction Center, The Online Tornado FAQ: http://www.spc.noaa.gov/faq/tornado/ Jan Null, Golden Gate Weather Services Tornado Page: http://ggweather.com/climate/tornado.htm Page last updated 2/11/05.