• Braun, S. A., , and Monteverdi J. P. , 1991: An analysis of a mesocyclone-induced tornado occurrence in northern California. Wea. Forecasting, 6 , 1331.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cole, R. O., , and Scanlon J. P. , 1955: Flood-producing rains in northern and central. California, December 16–26 1955. Mon. Wea. Rev., 83 , 336347.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Evans, J. S., , and Johns R. H. , 1995: A preliminary study of mountain tornadoes in Wyoming. Preprints, Seventh Conf. on Mountain Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 28–33.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1971: A proposed characterization of tornadoes and hurricanes by area and intensity. SMRP Paper 91, University of Chicago, 42 pp.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1989: The Teton–Yellowstone tornado of 21 July 1987. Mon. Wea. Rev., 117 , 19131940.

  • Grumm, R. H., , and Hart R. , 2001: Anticipating heavy rainfall events: Forecast aspects. Preprints, Symp. on Precipitation Extremes: Prediction, Impacts, and Responses, Albuquerque, NM, Amer. Meteor. Soc., 66–70.

    • Search Google Scholar
    • Export Citation
  • Hart, R., , and Grumm R. H. , 2001: Using normalized climatological anomalies to rank synoptic-scale events objectively. Mon. Wea. Rev., 129 , 24262442.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Junker, N. W., , Grumm R. H. , , Hart R. H. , , Bosart L. F. , , Bell K. M. , , and Pereira F. J. , 2008: Use of normalized anomaly fields to anticipate extreme rainfall in the mountains of northern California. Wea. Forecasting, 23 , 336356.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Junker, N. W., , Brennan M. J. , , Pereira F. , , Bodner M. J. , , and Grumm R. H. , 2009: Assessing the potential for rare precipitation events with standardized anomalies and ensemble guidance at the Hydrometeorological Prediction Center. Bull. Amer. Meteor. Soc., 90 , 445453.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kocin, P. J., , and Uccellini L. W. , 2004: A snowfall impact scale derived from northeast storm snowfall distributions. Bull. Amer. Meteor. Soc., 85 , 177194.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lussky, G. R., 1989: Heavy rains and flooding in Montana: A case for operational use of symmetric instability diagnosis. Wea. Forecasting, 4 , 186201.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lynott, R. E., , and Cramer O. P. , 1966: Detailed analysis of the 1962 Columbus Day windstorm in Oregon and Washington. Mon. Wea. Rev., 94 , 105117.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Monteverdi, J. P., , and Braun S. A. , 1988: An investigation of the 24 September 1986 “cold sector” tornado outbreak in northern California. National Weather Service Western Region Tech. Memo. NWS-203, 52 pp.

    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , Ralph F. M. , , White A. B. , , Kingsmill D. E. , , and Persson P. O. G. , 2002: The statistical relationship between upslope flow and rainfall in California’s Coastal Mountains: Observations during CALJET. Mon. Wea. Rev., 130 , 14681492.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , Ralph F. M. , , Wick G. A. , , Lundquist J. D. , , and Dettinger M. D. , 2008: Meteorological characteristics and overland precipitation impacts of atmospheric rivers affecting the west coast of North America based on eight years of SSM/I satellite observations. J. Hydrometeor., 9 , 2247.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Palmer, W. C., 1965: Meteorological drought. U.S. Weather Bureau Research Paper 45, 58 pp. [Available from NOAA Library and Information Services Division, Washington DC 20852].

    • Search Google Scholar
    • Export Citation
  • Simpson, R. H., 1974: The hurricane disaster potential scale. Weatherwise, 27 , 169186.

  • Stuart, N. A., , and Grumm R. H. , 2006: Using wind anomalies to forecast East Coast winter storms. Wea. Forecasting, 21 , 952968.

  • USDA, cited. 2008: Weekly Weather and Crop Bulletin. Vol 84, No. 50. [Available online at http://usda.mannlib.cornell.edu/usda/waob/weather_weekly/1990s/1997/weather_weekly-12-16-1997.pdf.

    • Search Google Scholar
    • Export Citation
  • Wagner, A. J., 1969: The weather and circulation of January 1969. Mon. Wea. Rev., 97 , 351358.

  • View in gallery

    Inset box representing the approximate domain utilized for the calculation of normalized anomalies.

  • View in gallery

    Contours of precipitable water (mm) and shading of the associated normalized anomaly (sigma) from 0000 UTC 17 Nov 1986.

  • View in gallery

    Contours of heights (m) and shaded normalized anomalies (sigma) from 0000 UTC 17 Nov 1986: (a) 500-hPa heights and anomaly, (b) 700-hPa heights and anomaly, (c) 850-hPa heights and anomaly, (d) mean SLP and anomaly.

  • View in gallery

    As in Fig. 3, but at 0000 UTC 18 Jul 1987.

  • View in gallery

    Wind barbs (kt) and shaded normalized anomalies (sigma) from 0000 UTC 18 Jul 1987: (a) 700-hPa winds and u-wind anomaly, (b) 700-hPa winds and υ-wind anomaly, (c) 850-hPa winds and u-wind anomaly, and (d) 850-hPa winds and υ-wind anomaly.

  • View in gallery

    As in Fig. 3, but at 0000 UTC 13 Oct 1962.

  • View in gallery

    As in Fig. 5, but at 0000 UTC 13 Oct 1962.

  • View in gallery

    Number of events with MTOTAL values ≥4 std dev by month.

  • View in gallery

    Average MTOTAL for top 10 events by month.

  • View in gallery

    Return period (months) for MTOTAL values (std dev).

  • View in gallery

    As in Fig. 10, but for MHEIGHT values (std dev).

  • View in gallery

    As in Fig. 10, but for MTEMP values (std dev).

  • View in gallery

    As in Fig. 10, but for MWIND values (std dev).

  • View in gallery

    As in Fig. 10, but for MMOIST values (std dev).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 28 28 4
PDF Downloads 23 23 0

Utilizing Normalized Anomalies to Assess Synoptic-Scale Weather Events in the Western United States

View More View Less
  • 1 National Weather Service, Salt Lake City, Utah
  • | 2 National Weather Service, State College, Pennsylvania
© Get Permissions
Full access

Abstract

Synoptic-scale weather events over the western United States are objectively ranked based on their associated tropospheric anomalies. Data from the NCEP 6-h reanalysis fields from 1948 to 2006 are compared to a 30-yr (1971–2000) reanalysis climatology. The relative rarity of an event is measured by the number of standard deviations that the 1000–200-hPa height, temperature, wind, and moisture fields depart from climatology. The top 20 synoptic-scale events were identified over the western United States, adjacent eastern Pacific Ocean, Mexico, and Canada. Events that composed the top 20 tended to be very anomalous in several, if not all four, of the atmospheric variables. The events included the northern Intermountain West region heavy rainfall and Yellowstone tornado of mid-July 1987 (ranked 5th), the Montana floods of September 1986 (ranked 4th), and the historic 1962 “Columbus Day” windstorm in the Pacific Northwest (ranked 10th). In addition, the top 10 most anomalous events were identified for each month and for each of the variables investigated revealing additional significant weather events.

Finally, anomaly return periods were computed for each variable at a variety of levels. To place a given anomaly in perspective for a specific level or element, forecasters need information on the frequency with which that anomaly is observed. These return periods can be utilized by forecasters to compare forecast anomalies to the actual occurrence of similar anomalies for the element and level of interest to gauge the potential significance of the event. It is believed that this approach may allow forecasters to better understand the historical significance of an event and provide additional information to the user community.

Corresponding author address: Randall A. Graham, National Weather Service, 2242 West North Temple, Salt Lake City, UT 84116. Email: randall.graham@noaa.gov

Abstract

Synoptic-scale weather events over the western United States are objectively ranked based on their associated tropospheric anomalies. Data from the NCEP 6-h reanalysis fields from 1948 to 2006 are compared to a 30-yr (1971–2000) reanalysis climatology. The relative rarity of an event is measured by the number of standard deviations that the 1000–200-hPa height, temperature, wind, and moisture fields depart from climatology. The top 20 synoptic-scale events were identified over the western United States, adjacent eastern Pacific Ocean, Mexico, and Canada. Events that composed the top 20 tended to be very anomalous in several, if not all four, of the atmospheric variables. The events included the northern Intermountain West region heavy rainfall and Yellowstone tornado of mid-July 1987 (ranked 5th), the Montana floods of September 1986 (ranked 4th), and the historic 1962 “Columbus Day” windstorm in the Pacific Northwest (ranked 10th). In addition, the top 10 most anomalous events were identified for each month and for each of the variables investigated revealing additional significant weather events.

Finally, anomaly return periods were computed for each variable at a variety of levels. To place a given anomaly in perspective for a specific level or element, forecasters need information on the frequency with which that anomaly is observed. These return periods can be utilized by forecasters to compare forecast anomalies to the actual occurrence of similar anomalies for the element and level of interest to gauge the potential significance of the event. It is believed that this approach may allow forecasters to better understand the historical significance of an event and provide additional information to the user community.

Corresponding author address: Randall A. Graham, National Weather Service, 2242 West North Temple, Salt Lake City, UT 84116. Email: randall.graham@noaa.gov

1. Introduction

Frequently, weather events that are deemed the most memorable or significant are those that depart substantially from the normal range of conditions experienced for a given location or a particular time of year, or those that have the biggest societal or economic impacts. Numerous methods have been developed to rank the severity and societal or economic impacts of meteorological events. The Fujita scale (Fujita 1971), the Saffir–Simpson scale (Simpson 1974), the Palmer drought index (Palmer 1965), and the Northeast snowfall impact scale (Kocin and Uccellini 2004) are a few scales developed to account for the severity of a weather event. Similarly, Hart and Grumm (2001, hereafter HG01) presented a ranking method to objectively assess synoptic-scale weather events utilizing tropospheric departures from normal. They noted that frequently the storms that are subjectively deemed the most significant are those that impact population centers and, therefore, receive the greatest media attention. When viewed objectively, through the use of anomalies, some well-known storms may prove to be not nearly as meteorologically rare as lesser-known storms that impacted areas of lower population. The methodology presented in HG01 attempts to utilize the normalized anomalies for a variety of elements and a series of levels to objectively rank synoptic events. However, the original work presented in HG01 only examined historical normalized anomalies for the eastern portion of North America. The work presented here applies a similar methodology as that presented in HG01, but the area of focus is instead centered on the western United States and its adjacent coastal waters (Fig. 1).

Based on normalized anomalies, all synoptic events from 1 January 1948 through 31 December 2006 were objectively ranked. Select historic events are presented based on their departures from normal and dominant anomalies. These data should facilitate the identification of historical weather events over western North America and, ultimately, may result in improved anticipation of future significant events.

2. Methodology

a. Datasets

Following the work of HG01, a comprehensive climatology was developed for the western United States in order to derive departures from normal for each 6-h period from January 1948 through December 2006. The National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis dataset (Kalnay et al. 1996) was utilized to generate this analysis. The NCEP–NCAR reanalysis dataset has a horizontal resolution of 2.5° × 2.5° and is available for 17 pressure levels. As in HG01, 10 of these levels were utilized for this study (Table 1). The climatology was developed at 6-h intervals for three basic variables over the range of 1000 to 200 hPa (height, temperature, and wind) as well as the full tropospheric value of precipitable water. The wind components were used and u- and υ-wind anomalies were computed. Climatological data for other variables, such as mean sea level pressure, were calculated as well. One significant difference between this study and HG01 is the use of precipitable water anomalies as opposed to the specific humidity anomalies. The reason for this difference is that several days of corrupt specific humidity data from February 1972 negatively impacted the specific humidity climatology and lowered the full tropospheric anomalies during the late winter (by yielding lower moisture departures during this period). This may partially explain the relative lull in significant events in the late winter found in HG01. The use of the precipitable water climatology eliminated this apparent bias in the western dataset.

Anomaly values were generated for over 85 000 six-hour time steps. The anomalies are relative to a climatology that is a subset of the entire period of investigation. The climatology was developed utilizing the 30-yr period extending from 1971 to 2000. With this climatology as a foundation, the ranking of the climatological departures was performed for the period from 0000 UTC 1 January 1948 through 1800 UTC 31 December 2006. It should be noted that HG01 used only 0000 and 1200 UTC data when developing rankings for eastern United States weather systems while 0000, 0600, 1200, and 1800 UTC data were utilized for this western U.S. review.

b. Definitions

Normalized anomalies were computed for the temperature, height, and u- and υ-wind components at each level (i.e., 1000–200 hPa) and for the full tropospheric value of precipitable water at all time periods. For each element, the value of MVARIABLE was computed, where MVARIABLE is the average departure (of all levels) from the climatological 21-day running mean in standard deviations. In other words, for a given variable (e.g., temperature) the greatest anomaly at each pressure level was identified and the absolute values of these anomalies were summed to generate the MVARIABLE (in this case, MTEMP). Variables used included temperature, heights, precipitable water, and the u- and υ-wind components. The overall departure, MTOTAL, was computed as the arithmetic average of the pressure-weighted variables (MTEMP, MHEIGHT, MMOIST, and MWIND) for each time step:
i1520-0434-25-2-428-eq1
For the determination of the MTOTAL, the MWIND value represents the vertical average of the u- or υ-wind anomalies, whichever was greatest. It should be noted that to arrive at an MTOTAL value, vertical continuity is not required in the summation of the MVARIABLE values as to account for system tilt with height. For example, the greatest moisture anomalies may have been present over the extreme southwest United States while, at the same time, the greatest wind anomalies may have been just off of the California coast. Despite the geographical separation between the two anomalies, they both would have contributed to the MTOTAL value for that particular time step. Therefore, there is only one MTOTAL calculated for each time step.

The most anomalous events were identified by the value of the MTOTAL. However, anomalies were also used to identify events relative to departures from normal for each variable. As such, events could be ranked by examining the MVARIABLE, also referred to as NTOTAL, to find the most significant events by variable (e.g., the top 10 MTEMP totals). For example, the top wind anomaly events could be examined independently of the top temperature or moisture events.

3. Results

The results are divided into two sections: rankings and return periods. The results are valid for the western half of the United States and the adjacent areas in the eastern Pacific (Fig. 1).

a. Rankings

The rankings for the normalized anomalies are broken down into several subsections. Section 3a(1) examines the top 20 MTOTAL departures during the 59-yr period, which should identify many of the largest and most significant storms to affect the region. Section 3a(2) examines the departures for each individual element (MHEIGHT, MTEMP, MWIND, and MMOIST) and section 3a(3) investigates MTOTAL rankings by month.

1) Top 20 normalized anomalies across the western United States from 1 January 1948 through 31 December 2006

The top 20 climatological anomalies (Table 2) represent the most significant multivariate synoptic-scale departures across the western United States for a period of nearly 60 yr. Several of these events are associated with memorable systems that had major societal impacts while others had little apparent economic or societal impact. A variety of event types were represented in the top 20 MTOTAL events for the study region, including coastal heavy rain, southwest U.S. heavy snow, severe weather outbreaks, and Pacific Northwest windstorms, among others.

When specific events are discussed, they are referred to relative to the date or time that the greatest MTOTAL value was analyzed. However, a single event may have produced very high MTOTAL values for several time periods and thus could have dominated the top events. Therefore, only the highest MTOTAL value was used for each event. For example, the top-ranked anomaly (4.497) in the dataset was analyzed at 0600 UTC 16 November 1986. However, this system exhibited significant MTOTAL values for several days as the upper low sat off of the southern California coast. Numerous time steps in the period from 15 November 1986 through 17 November 1986 would have ranked in the MTOTAL top 20 if more than one time step were allowed per storm. As such, the discussion of the events related to this particular system covers the period from 15 through 19 November 1986. Although not specifically stated, similar windows should be considered in the discussion of other historical events throughout the manuscript.

Since the MTOTAL calculation equally weights the impacts of the MHEIGHT, MTEMP, MWIND, and MMOIST values, a system generally needs to have large anomalies in several categories to rank in the top 20 of events overall. As a result, deep anomalous lows that exhibit significant departures in height, temperature, wind, and moisture are favored for inclusion in the top 20. Large anticyclones are poor candidates for top 20 events. Despite this tendency, or bias, toward strong cyclonic systems, there are several events in the top 20 that appear to have had a minimal impact on the western United States. Around 25% of the events in the top 20 were associated with cutoff lows off of southern California or the Baja coast. Several of these cutoff lows eventually moved inland and impacted the southwest United States while others remained offshore and were significant only as meteorological anomalies. As the MTOTAL approach is based on meteorological anomalies, it views an offshore cutoff low through the same lens as it views the devastating Columbus Day windstorm of 1962.

The magnitude of the top 20 events ranges from MTOTAL = 4.497 to MTOTAL = 4.029. These top 20 values over the western United States are composed of relatively lower MTOTAL values when compared to the values from the top 20 events across eastern North America, as discussed in HG01. In fact, the top event for the western United States would rank only sixth in the eastern North America dataset. To provide an additional perspective on the difference in the MTOTAL values between the domain in HG01 and the western domain presented here, the 20th-ranked event in HG01 (22 January 1959 with an MTOTAL of 4.176) would have ranked ninth in the top 20 events for the western domain. The discrepancy is even greater when the western analysis is completed using specific humidity for MMOIST, as was done in HG01. When this is done, the 20th-ranked event in the eastern domain would rank 5th in the western domain.

The reasons for the higher MTOTAL values in the eastern domain are not immediately clear, but a simple comparison of the top events for each variable yields some potential insight. It is clear that MMOIST differences are a key factor in the MTOTAL differences between the eastern and western domains. When the western analysis is performed utilizing the specific humidity for the MMOIST, the 10th-ranked event in the east would rank as the fourth largest MMOIST anomaly in the west. It is believed that this difference is primarily a result of exposure to more significant moisture surges associated with warmer sea surface temperatures (SSTs) in the Gulf of Mexico and the Gulf Stream while cooler water off of the west coast and complex terrain may result in less extreme moisture intrusions in the western United States. Additionally, it also appears that both MTEMP and MWIND play a role in the differences in the magnitude of the MTOTAL between the two domains. The 10th-ranked MTEMP event in the eastern domain would rank fourth in the west while the eastern domain has three MWIND events with values in excess of five standard deviations and the western domain has none. It is believed that the continental impacts are much greater in the eastern domain, resulting in extremes in heat and cold while these influences are tempered in the west due to the prevailing flow off of the Pacific Ocean. The increased baroclinicity associated with the significant events in the eastern domain may also play a role in the MWIND differences as well. The top 10 MHEIGHT departures were comparable between the two domains, indicating that the reasons for the larger MTOTAL values in the eastern domain may largely be determined by differences in MTEMP, MMOIST, and MWIND. While these simple comparisons offer some plausible reasons for the MTOTAL differences noted between the two domains, a more rigorous investigation would very likely yield additional insight.

Finally, the impact of complex terrain on landfalling systems in the western United States cannot be ignored. Systems that impinge upon the western United States tend to be greatly modified through their interaction with the complex topography, resulting in less coherent storm systems in the interior west. Junker et al. (2008, 2009) have demonstrated that strong flow of warm moist air from the Pacific can produce record precipitation events in the absence of intense surface and midlevel cyclones. Thus, in the western United States, the extreme terrain may be a critical factor in many significant weather events.

Despite the tendency for lower MTOTAL values in the western domain, impressive events still compose much of the top 20 list for the western United States. Of note are events such as the Columbus Day storm of 1962 (Lynott and Cramer 1966), a significant Pacific Northwest severe weather outbreak in May 2006, and a July 1987 event, which included an F4 tornado in Yellowstone National Park (Fujita 1989; Evans and Johns 1995). Despite these impressive events, around one-quarter of the systems that compose the top 20 are not known to have a major societal or economic impact across the western United States. Several of these anomalous events, as meteorologically significant as they appeared, were relatively low-impact events, and as previously stated, many were associated with cutoff lows off the coast of the southwest United States.

The most anomalous multivariate synoptic-scale event across the western United States for the period from 1948 through 2006 was associated with a cutoff low on 16 November 1986. The system was quite anomalous with respect to heights, winds, and precipitable water values. For most of the period when it was at its most anomalous, the system remained offshore of southern California. On 18 November, the upper low became an open wave and moved into the southwestern United States. Localized heavy rain was reported with numerous sites recording rainfall totals of 1–2 in. (25–50 mm) from southern California into southwest Colorado and western New Mexico. Numerous sites recorded precipitation amounts that were among the wettest November days on record, including Lompoc, California [2.01 in. (51 mm); the second wettest November day on record]; Monument Valley, Utah [1.06 in. (27 mm); the wettest November day on record); Shoshone, Arizona [1.00 in. (25 mm); the second wettest November day on record], and Desert Rock, Nevada [0.87 in. (22 mm); the second wettest November day on record]. San Diego, California, recorded 1.18 in. (30 mm) of rain, which exceeded the average November rainfall in a single day. The heavy rain in southern California resulted in flooding along the San Diego River. In addition, this system was associated with two tornadoes in the Phoenix, Arizona, area, accounting for two of only nine November tornadoes on record in the state.

This event stood out in that it supported departures that ranked in the top 10 with respect to height (fifth overall), wind (sixth overall), and moisture (fifth overall) when compared to the normalized climatology. The most significant departure associated with this system was the precipitable water anomaly, which was greater than seven standard deviations above normal (Fig. 2). Height anomalies near five standard deviations (Fig. 3) below normal were associated with the cutoff low as it sat off of the coast of California. Significant MWIND totals were observed as well with anomaly values in excess of four standard deviations in both the u- and υ-wind components at various levels (not shown). The degree of departure in three of the four elements investigated is what makes this event truly stand out from all of the other events in the 59-yr dataset. However, despite the significant weather associated with this system, it is not an event that was extensively documented.

In contrast, a July 1987 event, one of the most memorable events on record in the western United States, came in as the fifth-most anomalous event in the period from 1948 to 2006. The impacts from this system were most pronounced from the northern Great Basin to the northern Intermountain West region and into the plains of central and eastern Montana. The storm moved through the region in the middle of July 1987 and is best known for being associated with an F4 tornado that struck Yellowstone National Park on 21 July. While this event is well documented due to the severe weather outbreak that stretched from northern Utah into northwest Wyoming, and in particular the F4 tornado in Yellowstone National Park, it also had far-reaching impacts to the north. Widespread rainfall totals in excess of 2 in. (50 mm), with some locations receiving over 4 in. (100 mm), were reported across Idaho and Montana on 16–17 July, with some sites receiving warm season (April–September) record 24-h precipitation amounts (i.e., Summer Lake, Oregon, with 1.67 in. on 17 July). In addition, Crater Lake National Park Headquarters reported an inch of snow on 18 July, which is the latest in the season that the site has recorded an inch of snow or greater (1919–2007). While heavy precipitation in the warm season in this region is not, in itself, particularly unusual, the widespread coverage of the heavy precipitation was impressive.

Similar to the top event in the dataset (16 November 1986), this storm exhibited impressive departures that ranked in the top 10 in three of the four elements investigated (height, temperature, and wind). On the day of the greatest MTOTAL for this event, 18 July 1987, or on the day prior, anomalies for height, temperature, and wind ranked in the top four for all events dating back to 1948. Height anomalies in excess of five standard deviations (Fig. 4) below normal were associated with the unseasonably deep trough as it moved into the Great Basin and, in particular, across southern California and Nevada. Anomaly values in excess of four standard deviations from normal were associated with both the u and υ winds at various levels (Fig. 5), resulting in highly anomalous MWIND totals. Even though the MMOIST total for this event was not in the top 10 for all moisture anomalies, it still supported significant moisture anomalies, with precipitable water values between two to three standard deviations above normal.

One of the most significant storms in the western United States during the twentieth century was the Columbus Day storm of 1962. It is often referred to as the “Storm of the Century” for the western United States and as such is worthy of additional review given its place in weather lore across the west. The Columbus Day storm ranked 10th in the top 20 MTOTAL events for the western United States with a value of 4.142 as analyzed on 13 October 1962 (Table 2). The damage associated with this event remains unequaled in the historical records of the Pacific Northwest and is the standard by which all other Pacific Northwest windstorms are measured. The most intense winds with the storm were recorded on 11 October with 145 mi h−1 measured at Cape Blanco, Oregon. The actual gusts were likely higher than this as the anemometer failed at that point (Lynott and Cramer 1966). There were numerous other gusts well over 100 mi h−1, including 130 mi h−1 at Mount Hebo Air Force Base (the maximum reading available); 127 mi h−1 at Corvallis, Oregon, before the anemometer was destroyed; and Newport, Oregon, with 138 mi h−1. In Oregon and Washington alone, 15–17 million board-acre feet of timber were felled. The storm resulted in an estimated $230–280 million (in 1962 dollars) in property damage from California north to British Columbia, Canada, with the worst of the damage occurring in Oregon. Impressive height anomalies were associated with this historic system with low-level (e.g., SLP, 850 hPa, 700 hPa, etc.) analyzed departures of at least five standard deviations below normal (Fig. 6). In addition, low-level υ-wind anomalies were in excess of five standard deviations above normal as the core of intense winds moved inland (Fig. 7). It should be noted that that the MHEIGHT anomalies from this event ranked in the top 6 (Table 3a) and MWIND anomalies in the top 4 (Table 3c) for all events dating back to 1948. However, the event lacked extreme thermal and moisture anomalies, which kept the event from being ranked in the overall top 5.

2) Top 10 anomalies by variable

In the previous section, the MTOTAL for each 6-h period for the 59-yr span was used to objectively identify the most anomalous weather systems. In this section, the individual vertically integrated variables for height, temperature, wind, and precipitable water are used to examine the events that had the largest anomalies by parameter. The top 10 anomalies for each of these variables are listed in Tables 3a –d.

(i) MHEIGHT anomalies

The largest MHEIGHT anomaly of 6.884 (Table 3a) was associated with a large cutoff low situated off of the Baja coast on 10 November 1969. The second largest MHEIGHT anomaly (5.619) over the 59-yr period occurred on 9 June 1976. On 9 June 1976 an unseasonably strong trough sat off of the southern California coast. As this system moved toward the north-central plains, it was associated with a significant severe weather outbreak across North and South Dakota. The 11 June outbreak resulted in the most tornadoes recorded in a single day in North Dakota (nearly 30 separate tornado events) and produced large hail and flash flooding in the Black Hills of South Dakota. The majority of the remaining MHEIGHT top 10 events were associated with heavy precipitation events while a few were associated with cutoff lows off the Baja coast.

(ii) MTEMP anomalies

The largest value of MTEMP (5.063; see Table 3b) was associated with a sprawling area of high pressure, which settled over the Great Basin and extended into southern Texas on 13 December 1997. As the area of high pressure moved south, 700-hPa temperatures dropped to as low as −16°C across southern Texas and northern Mexico. Frigid air settled into much of the Great Basin and the southwest United States, and the system was associated with a widespread freeze across southern Texas with temperatures even dropping to near or below freezing in the Phoenix area. Morning lows in Yellowstone, Wyoming, dropped as low as −32°F (−36°C). Additionally, a rare sleet event occurred in Brownsville, Texas, in what was only the 12th occurrence of frozen precipitation in 50 yr (USDA 2008). As impressive as the cold air was over the Great Basin, the largest temperature anomalies were observed over northern Mexico. This resulted in a rare cold episode over northern Mexico and snowfall reportedly paralyzed the region and was blamed for the deaths of 12 people. The city of Guadalajara, Mexico, reported snow for the first time since 1881. The second largest MTEMP anomaly (4.958) was associated with the 18 July 1987 western U.S. trough, which was also responsible for the fifth largest MTOTAL departure on record in the western United States [see section 3a(1)].

(iii) MWIND anomalies

The largest MWIND value (4.962; see Table 3c) occurred on 2 July 1997 in association with a vigorous trough moving onto the California coast. The system resulted in local precipitation amounts in excess of 1 in. (25 mm) across northern California and Oregon on 30 June and 1 July 1997. Heavy precipitation spread to the northeast with flooding rains reported in eastern Montana on 1–2 July 1997. Several sites set all-time single-day precipitation records [Lindsay, Montana, 5.52 in. (140 mm); Terry, Montana, 4.68 in. (118 mm)] and amounts of nearly 6 in. (152 mm) around Sidney, Montana, resulted in a dam failure. The second largest MWIND anomaly (4.701) occurred on 26 June 1996. A deep trough moved down the West Coast and then moved inland with a 110-kt southerly jet at 250 hPa on 26 June 1996, increasing to 130 kt over the Great Basin on 27 June. As the system moved into the Great Basin, it was associated with a significant severe weather outbreak, with hail up to softball size, across western and central Montana on 26 June. Several of the top 10 MWIND events were associated with heavy rain events (2 July 1997, 26 June 1969, 22 August 1968), two with severe weather outbreaks (26 June 1996 and 17 July 1987), one with a synoptic-scale windstorm in the Pacific Northwest (13 October 1962), one with a significant winter storm (18 January 1988), and one with a downslope windstorm in Washington (24 December 1983).

(iv) MMOIST anomalies

The top MMOIST anomaly (8.479; see Table 3d) occurred on 18 May 1956. A plume of deep moisture lifted into Baja California with precipitable water values around 2 in. (50 mm) noted off of the southern California coast. This moisture anomaly was initially associated with precipitation in California although amounts were largely less than 0.5 in. (13 mm). As this moisture plume lifted north, Medford, Oregon, received 1.67 in. of precipitation on 18 May 1956, which is the greatest single-day total on record (1928–2008) between 1 May and 31 August. Despite the strength of the anomaly, the areal extent of the precipitation was surprisingly modest, although it is worth noting that the significant moisture anomaly was largely confined to the Baja California region and far western Mexico. The remainder of the MMOIST top 10 was largely composed of significant precipitation events. This includes the California and southern Arizona heavy rain and northern Arizona snow event of 22–23 March 1954, which was associated with the second-highest MMOIST values. Numerous storm rainfall totals of 10–12 in. (250–300 mm) were reported in the California mountains and 32 in. (81 cm) of snow fell in Bright Angel, Arizona, in a 24-h period. Several other record-breaking precipitation events are included in the top 10, including those on 31 March 1986 and 23 November 1965, both of which included widespread heavy precipitation across the southwest United States.

3) Top 10 anomalies by month

In addition to the top 10 anomalies by variable, the top 10 events were computed by month and compiled in Tables 4a –4l. The monthly data revealed additional historic western U.S. weather events and provided a seasonally adjusted set of events, as many of these monthly events were not included in the MTOTAL or MVARIABLE tables. Many of these events produced significant weather and were worthy of inclusion in this study as the goal is to aid in detecting significant weather events.

Several events identified in the monthly top 10s include the historic heavy rain and flood event in California in early January 1997 (Junker et al. 2008) and the extreme cold snap and heavy snow event of 10–11 January 1949, which produced measurable snow in Los Angeles, California, and Las Vegas, Nevada. The monthly data also revealed the 3 September 1961 event, which included a record early season snowfall in Denver, Colorado.

Some interesting artifacts of the climatology are revealed upon further investigation of the monthly top 10 lists. None of the western U.S. top 10 January events made the overall top 20 events while each of the January top 10 events made the overall top 20 in HG01. The month that was most heavily represented in the MTOTAL top 20 was November with five events, followed by March, April, and May with three events each. While the eastern U.S. top 20 was dominated by winter weather events, the western U.S. climatology was heavily influenced by transition season events with 14 of the top 20 events occurring in the months of March, April, May, and November.

In an effort to assess the seasonality of significantly anomalous events, the number of events with MTOTAL values of 4 or greater was calculated for each month (Fig. 8). It was found that events with MTOTAL of 4 or greater most frequently occurred in the transition periods of spring and fall. The HG01 dataset was dominated by significant winter storm events while the western dataset presented here exhibits more variety in the event types that comprise the most anomalous events. The tendency for more significant departures in the spring and fall in the western U.S. domain is also noted in the average MTOTAL for the top 10 events for each month (Fig. 9). The average values for the top 10 events in the spring and fall months clearly stand out when compared to the remainder of the year.

b. Return periods

In addition to objectively ranking these synoptic-scale events, the ultimate goal of this work is to utilize anomaly information in the forecast process across the western United States. To place a given anomaly value in perspective for a specific element and level, or an M value, it is useful to know how frequently that anomaly is observed within the domain. For example, based on the anomalies analyzed for this study, the return period for a υ-wind anomaly of at least +4.5 standard deviations at 700 hPa occurring anywhere in the western United States is about 6 months. In other words, a υ-wind anomaly of this magnitude, or greater, at 700 hPa was observed a little over 100 times in the over 85 000 six-hour time steps included in the dataset. This type of information should provide forecasters with a better sense of the actual rarity or, conversely, the insignificance of an observed or forecast anomaly.

To quantify this information, return periods have been calculated for different elements and levels. Bins with a width of 0.1 standard deviations for each 6-h period were utilized to assess the frequency of occurrence of observed MTOTAL, MHEIGHT, MTEMP, MWIND, and MMOIST values. The return periods are related to the monthly frequency with which a given anomaly is observed (Figs. 10 –14).

The return periods for MTOTAL values are displayed in Fig. 10. The most common MTOTAL value in the database is 2.3, which, on average, is observed in more than 13 six-hour time steps per month. To expand on this, MTOTAL values between 2.1 and 2.7 are observed numerous times in a given month (more than 8 times per month) and thus are quite common. Meanwhile, MTOTAL values of around 1.7 and 3.5 are observed about once a month. Values of MTOTAL that are lower than 1.7 and greater than 3.5 occur with rapidly decreasing frequency as you move away from those two values. In other words, it is just as rare to have a very low MTOTAL value as it is to have one that is significantly large (e.g., greater than four standard deviations). The MTOTAL of 0.9 standard deviations represents the “least active” weather found in the database, having been observed once in the 59-yr period of the study. This value is much lower than the minimum value observed in HG01, which was 1.3 standard deviations. In general, MTOTAL values exceeding 3.7 standard deviations would warrant closer scrutiny from a forecast perspective as MTOTAL values this high are only observed a small number of times each year. For comparison purposes, the return periods for MHEIGHT, MTEMP, MWIND, and MMOIST are shown in Figs. 11 –14.

Return periods were also calculated for numerous levels (e.g., 925, 850, 700 hPa, etc.) for four of the five primary variables (i.e., heights, temperature, υ- and u-wind components). From a forecast perspective, it is likely that the return periods for specific fields and levels will prove more useful than MTOTAL return periods in that forecasters will be able to focus on the truly anomalous elements and levels when making a forecast. Charts for these anomaly return periods are available online (http://www.wrh.noaa.gov/slc/projects/anomalies/return_intervals/index.htm).

4. Forecasting implications

It is clear that the use of normalized anomalies and return periods can place events into historical perspective, but normalized anomalies can also be beneficial in the real-time forecast process. In recent years, the use of anomalies in the forecast process has steadily increased and studies examining the use of anomalies to forecast a variety of significant weather events, including East Coast winter storms (Stuart and Grumm 2006), extreme mountain rainfall in northern California (Junker et al. 2008), and heavy rainfall events (Grumm and Hart 2001), have appeared in the literature. It is believed that the interrogation of specific elements (e.g., u or υ wind) and specific levels (e.g., 850 and 700 hPa, etc.) provides more utility than the examination of MTOTAL values in a real-time forecast environment. Looking at the maximum MTOTAL values associated with an event helps assess the rarity of an event. However, the MTOTAL value does not provide the forecaster with critical information as to which elements and levels are most anomalous, thereby limiting its utility as a real-time forecast tool. Forecasters are instead encouraged to interrogate significant anomalies by element and level.

The southern California heavy rain event of 18–28 January 1969 is an example of the utility of interrogating critical elements and levels pertinent to specific event types. Record-setting rainfall totals were recorded across portions of southern California during this period (Wagner 1969) with a few mountain sites reporting in excess of 30 in. (762 mm) of rain during this 10-day period, including a total of 41.88 in. (1064 mm) at Lake Arrowhead, California. Widespread flooding was associated with this significant rainfall event along California’s Central Valley rivers, which led to the reformation of Tulare Lake in the San Joaquin Valley. However, this historic rainfall and flood event does not appear in the MTOTAL top 20, with 3.212 being the largest MTOTAL analyzed in the western United States during this 10-day period. An investigation of this event reveals that there were indeed significant anomalies associated with this event, but the large anomalies were confined to only a couple of elements (primarily precipitable water and winds) and were largely confined to the lower portion of the troposphere (i.e., below 700 hPa). The precipitable water anomalies associated with a series of landfalling atmospheric rivers (Neiman et al. 2008) were between 3.5 and 4.5 standard deviations above normal for the two periods of significant precipitation during this event. In addition, low-level wind (both the u- and υ-wind components) anomalies were between +3.0 and +4.5 standard deviations for significant stretches of time during this event with the 925-hPa υ-wind anomalies exceeding +5 standard deviations for a period on 25 January. However, the upper-level wind anomalies were more modest, weighting the MWIND values down a bit. Similarly, the greatest height and temperature anomalies (those in excess of 3 standard deviations) were largely confined to the lower levels. The result is that the MTOTAL values for this event were generally between 2.5 and 3.0 standard deviations for most of the event. However, significant anomalies were present in fields (i.e., moisture and wind) and levels (i.e., below 700 hPa) that are critical for forecasting this particular type of event, as discussed in Neiman et al. (2002) and Junker et al. (2008). This implies that some event types, such as coastal heavy rainfall events, can occur with significant anomalies present only in critical fields while the MTOTAL remains relatively modest.

It is believed that forecasters could utilize output from an Ensemble Prediction System (EPS) or deterministic medium-range model (e.g., the Global Forecast System, GFS) to interrogate forecast anomalies for elements (e.g., precipitable water, temperature, u-wind component) and at levels (e.g., 850 and 700 hPa) pertinent to this specific event type. The heavy rain event of January 1969 suggests the need to understand the key anomalies that may contribute to forecasting specific event types (Junker et al. 2009). This information could serve as an excellent forecast tool in real time for forecasters attempting to anticipate the significance of a pending event.

5. Conclusions

Utilizing the methodology presented in HG01, a climatology of normalized anomalies was developed for the western United States for the period of 1948–2006. Six-hourly anomaly departures for a series of levels were examined for height, temperature, wind, and moisture from 1948 through 2006. In addition, the total anomaly (accounting for the MHEIGHT, MTEMP, MWIND, and MMOIST) was identified for each 6-h time step. Through this process, the most meteorologically anomalous periods for the western United States between 1948 and 2006 were identified.

Similar to the results found for the eastern United States, the larger MTOTAL events in the western United States were dominated by significant weather events, some of which appeared in the published literature. However, a smaller percentage of the top 20 events in the western U.S. domain appeared in the published literature when compared to the events in HG01. This is likely a result of the lower population base in the western United States. It should be clear that there is not a linear relationship between the presence of significant anomalies and the associated human impacts as location, time of year, time of day, etc. all play a role in the observed impacts. Therefore, not all events that exhibit big anomalies will make the news or be the subject of focused research. However, by examining Storm Data entries, newspaper articles, National Climatic Datat Center (NCDC) records, etc., it is clear that many of these anomalous events in the western United States were indeed associated with unique and significant weather. In several cases, events that initially appeared to have had a minimal impact were found, upon further investigation, to have been significant events given the climatology of the area in which they occurred. Events listed in Table 2 produced a wide spectrum of significant weather ranging from a rare F4 tornado event to record snow events, extreme rainfall and flooding, and significant windstorms.

Interestingly, many of the events in the top 20 tended to be anomalous with respect to all four variables investigated (temperatures, heights, moisture, and winds). Some events, such as the Columbus Day storm, were very anomalous with respect to heights (Fig. 6, Table 3a) and winds (Fig. 7, Table 3c) but lacked significant thermal and moisture anomalies. The MTEMP for this event was 2.397, which was the lowest MTEMP value for any top 20 event. The MTOTAL concept may be biased toward events that can bring together a wide range of meteorological conditions producing strong anomalies in all four variables. Clearly, the Columbus Day storm of 1962, with a mean SLP anomaly (−8.172 standard deviations), was a significant weather event. The −8.172 standard deviation anomaly was the largest SLP anomaly observed in the western United States during the study period. This was comparable to the anomalies associated with the deep cyclones in the eastern United States. It should be noted that despite the lack of large thermal and moisture anomalies, the Columbus Day storm of 1962 is the most anomalous October event in the western United States.

Examining the data by variable (Tables 3) suggested that many of the top 10 MVARIABLE events could also be top 20 MTOTAL events. It was also noted that many of the large height anomalies were associated with heavy rain and snow events. Virtually all of the anomalous precipitable water events were associated with significant heavy rain events. Interestingly, the top 10 MTEMP anomalies were generally associated with significant cold spells or heavy precipitation events. The absence of significant western U.S. heat waves in the top 10 indicates that MTEMP values may not be an appropriate approach for anticipating heat waves in the west. Investigation of several western U.S. heat waves (e.g., July 2006, July 2005, July 2002, etc.) indicates that positive thermal anomalies (say, standard deviations of +2 to +3) at 850 and 700 hPa are better indicators of the potential for extreme heat than is a full tropospheric approach using MTEMP values. The top 10 wind anomaly events appeared to be associated with heavy rain and severe weather events. The presence of strong winds increases the likelihood of high shear, a known ingredient in producing strong and long-lived convection. As these weather systems move into the eastern side of the domain, they likely are capable of tapping low-level moisture, adding to the instability favoring severe weather. Farther west in the domain, strong winds are often associated with significant moisture flux in active storm environments and this is known to be correlated with enhanced orographically forced precipitation (Junker et al. 2008).

Examining the monthly top 10 events revealed some trends in the data and contrasts with eastern North American events. For example, in HG01, the top 10 January events were all top 20 MTOTAL events, suggesting a bias toward strong winter storms dominating the larger-scale events. However, in the western United States none of the top 10 January events was represented in the top 20 MTOTAL events. The average MTOTAL value for the eastern U.S. January top 10 was 4.439 compared to 4.191 in the western United States. The lack of significant baroclinicity along the West Coast relative to the strong baroclinicity along the East Coast may play a role in this striking lack of top 10 January events making top 20 MTOTAL events. This lack of deep thermal anomalies was certainly present in the Columbus Day windstorm.

Ultimately, it is the goal of this work to set the stage for better anticipation of significant and historical events. It is hoped that the development of the anomaly return periods will allow forecasters to assess model output and place the forecast anomalies into perspective with respect to the relative frequency of their occurrence. Further understanding of the anomalies associated with significant historical storms may also enable forecasters to better anticipate future events. Finally, investigation of anomalies associated with specific event types (e.g., Pacific Northwest windstorms, western U.S. heat waves, coastal heavy rain events, etc.) may yield additional clues as to the tropospheric departures that are pertinent for forecasting these events. Through the examination of the analyzed anomalies, as well as numerous significant historical events, it is clear that the lack of significant synoptic-scale anomalies does not preclude a high-impact event from occurring on a smaller scale. However, the presence of large anomalies is generally associated with significant weather events although they may not have substantial societal or economic impacts.

Acknowledgments

The authors acknowledge Robert Hart for his insights and contributions related to examining synoptic scale anomalies. The authors also thank Larry Dunn (Meteorologist-in-Charge, National Weather Service, Salt Lake City) for his support and encouragement of this research. We also recognize Monica Traphagan (National Weather Service, Salt Lake City) for her assistance in the generation of the anomaly return periods and Lisa Verzella and Michael Olson (University of Utah) for their assistance in examining past events. Finally, we greatly appreciate the constructive feedback provided by two anonymous reviewers.

REFERENCES

  • Braun, S. A., , and Monteverdi J. P. , 1991: An analysis of a mesocyclone-induced tornado occurrence in northern California. Wea. Forecasting, 6 , 1331.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cole, R. O., , and Scanlon J. P. , 1955: Flood-producing rains in northern and central. California, December 16–26 1955. Mon. Wea. Rev., 83 , 336347.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Evans, J. S., , and Johns R. H. , 1995: A preliminary study of mountain tornadoes in Wyoming. Preprints, Seventh Conf. on Mountain Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 28–33.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1971: A proposed characterization of tornadoes and hurricanes by area and intensity. SMRP Paper 91, University of Chicago, 42 pp.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1989: The Teton–Yellowstone tornado of 21 July 1987. Mon. Wea. Rev., 117 , 19131940.

  • Grumm, R. H., , and Hart R. , 2001: Anticipating heavy rainfall events: Forecast aspects. Preprints, Symp. on Precipitation Extremes: Prediction, Impacts, and Responses, Albuquerque, NM, Amer. Meteor. Soc., 66–70.

    • Search Google Scholar
    • Export Citation
  • Hart, R., , and Grumm R. H. , 2001: Using normalized climatological anomalies to rank synoptic-scale events objectively. Mon. Wea. Rev., 129 , 24262442.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Junker, N. W., , Grumm R. H. , , Hart R. H. , , Bosart L. F. , , Bell K. M. , , and Pereira F. J. , 2008: Use of normalized anomaly fields to anticipate extreme rainfall in the mountains of northern California. Wea. Forecasting, 23 , 336356.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Junker, N. W., , Brennan M. J. , , Pereira F. , , Bodner M. J. , , and Grumm R. H. , 2009: Assessing the potential for rare precipitation events with standardized anomalies and ensemble guidance at the Hydrometeorological Prediction Center. Bull. Amer. Meteor. Soc., 90 , 445453.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kocin, P. J., , and Uccellini L. W. , 2004: A snowfall impact scale derived from northeast storm snowfall distributions. Bull. Amer. Meteor. Soc., 85 , 177194.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lussky, G. R., 1989: Heavy rains and flooding in Montana: A case for operational use of symmetric instability diagnosis. Wea. Forecasting, 4 , 186201.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lynott, R. E., , and Cramer O. P. , 1966: Detailed analysis of the 1962 Columbus Day windstorm in Oregon and Washington. Mon. Wea. Rev., 94 , 105117.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Monteverdi, J. P., , and Braun S. A. , 1988: An investigation of the 24 September 1986 “cold sector” tornado outbreak in northern California. National Weather Service Western Region Tech. Memo. NWS-203, 52 pp.

    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , Ralph F. M. , , White A. B. , , Kingsmill D. E. , , and Persson P. O. G. , 2002: The statistical relationship between upslope flow and rainfall in California’s Coastal Mountains: Observations during CALJET. Mon. Wea. Rev., 130 , 14681492.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , Ralph F. M. , , Wick G. A. , , Lundquist J. D. , , and Dettinger M. D. , 2008: Meteorological characteristics and overland precipitation impacts of atmospheric rivers affecting the west coast of North America based on eight years of SSM/I satellite observations. J. Hydrometeor., 9 , 2247.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Palmer, W. C., 1965: Meteorological drought. U.S. Weather Bureau Research Paper 45, 58 pp. [Available from NOAA Library and Information Services Division, Washington DC 20852].

    • Search Google Scholar
    • Export Citation
  • Simpson, R. H., 1974: The hurricane disaster potential scale. Weatherwise, 27 , 169186.

  • Stuart, N. A., , and Grumm R. H. , 2006: Using wind anomalies to forecast East Coast winter storms. Wea. Forecasting, 21 , 952968.

  • USDA, cited. 2008: Weekly Weather and Crop Bulletin. Vol 84, No. 50. [Available online at http://usda.mannlib.cornell.edu/usda/waob/weather_weekly/1990s/1997/weather_weekly-12-16-1997.pdf.

    • Search Google Scholar
    • Export Citation
  • Wagner, A. J., 1969: The weather and circulation of January 1969. Mon. Wea. Rev., 97 , 351358.

Fig. 1.
Fig. 1.

Inset box representing the approximate domain utilized for the calculation of normalized anomalies.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 2.
Fig. 2.

Contours of precipitable water (mm) and shading of the associated normalized anomaly (sigma) from 0000 UTC 17 Nov 1986.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 3.
Fig. 3.

Contours of heights (m) and shaded normalized anomalies (sigma) from 0000 UTC 17 Nov 1986: (a) 500-hPa heights and anomaly, (b) 700-hPa heights and anomaly, (c) 850-hPa heights and anomaly, (d) mean SLP and anomaly.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 4.
Fig. 4.

As in Fig. 3, but at 0000 UTC 18 Jul 1987.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 5.
Fig. 5.

Wind barbs (kt) and shaded normalized anomalies (sigma) from 0000 UTC 18 Jul 1987: (a) 700-hPa winds and u-wind anomaly, (b) 700-hPa winds and υ-wind anomaly, (c) 850-hPa winds and u-wind anomaly, and (d) 850-hPa winds and υ-wind anomaly.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 6.
Fig. 6.

As in Fig. 3, but at 0000 UTC 13 Oct 1962.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 7.
Fig. 7.

As in Fig. 5, but at 0000 UTC 13 Oct 1962.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 8.
Fig. 8.

Number of events with MTOTAL values ≥4 std dev by month.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 9.
Fig. 9.

Average MTOTAL for top 10 events by month.

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 10.
Fig. 10.

Return period (months) for MTOTAL values (std dev).

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 11.
Fig. 11.

As in Fig. 10, but for MHEIGHT values (std dev).

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 12.
Fig. 12.

As in Fig. 10, but for MTEMP values (std dev).

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 13.
Fig. 13.

As in Fig. 10, but for MWIND values (std dev).

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Fig. 14.
Fig. 14.

As in Fig. 10, but for MMOIST values (std dev).

Citation: Weather and Forecasting 25, 2; 10.1175/2009WAF2222273.1

Table 1.

Pressure levels utilized in the determination of normalized anomalies by element. Here, MVARIABLE represents full tropospheric representation.

Table 1.
Table 2.

Top 20 normalized MTOTAL events for the period 1948–2006.

Table 2.

Table 3a. Top 10 normalized MHEIGHT events for the period 1948–2006.

i1520-0434-25-2-428-t03a

Table 3b. Top 10 normalized MTEMP events for the period 1948–2006.

i1520-0434-25-2-428-t03b

Table 3c. Top 10 normalized MWIND events for the period 1948–2006.

i1520-0434-25-2-428-t03c

Table 3d. Top 10 normalized MMOIST events for the period 1948–2006.

i1520-0434-25-2-428-t03d

Table 4a. Top 10 January normalized departures from climatology.

i1520-0434-25-2-428-t04a

Table 4b. Top 10 February normalized departures from climatology.

i1520-0434-25-2-428-t04b

Table 4c. Top 10 March normalized departures from climatology.

i1520-0434-25-2-428-t04c

Table 4d. Top 10 April normalized departures from climatology.

i1520-0434-25-2-428-t04d

Table 4e. Top 10 May normalized departures from climatology.

i1520-0434-25-2-428-t04e

Table 4f. Top 10 June normalized departures from climatology.

i1520-0434-25-2-428-t04f

Table 4g. Top 10 July normalized departures from climatology.

i1520-0434-25-2-428-t04g

Table 4h. Top 10 August normalized departures from climatology.

i1520-0434-25-2-428-t04h

Table 4i. Top 10 September normalized departures from climatology.

i1520-0434-25-2-428-t04i

Table 4j. Top 10 October normalized departures from climatology.

i1520-0434-25-2-428-t04j

Table 4k. Top 10 November normalized departures from climatology.

i1520-0434-25-2-428-t04k

Table 4l. Top 10 December normalized departures from climatology.

i1520-0434-25-2-428-t04l
Save