WESTERN REGION TECHNICAL ATTACHMENT
NO. 02-04
FEBRUARY 5, 2002

SIGNIFICANT SNOW IN THE SIERRA NEVADA FROM
MARITIME POLAR ORIGINS

Alexander Tardy, Weather Forecast Office, Sacramento, CA

Introduction

The Sierra Nevada is well known for its heavy snowfalls dating back to the infamous blizzard of the 1840's which stranded the Donner Party in the rugged terrain (McLaughlin 2001). The name Sierra Nevada implies snow since its Spanish translation is the snow covered (Nevada) mountains (Sierra). Annual snowfall ranges from 400 to 450 in (1010 to 1150 cm) across most locations. The mountain range extends nearly 400 miles (645 km) in length over the eastern part of California, and peaks reach up to 14,000 ft MSL.

Forecasting snowfall across the Sierra Nevada is shared between three National Weather Service (NWS) offices. The office in Sacramento is responsible for forecasts over the western slopes of the mountain range up to the Sierra Nevada crest (e.g., up to Donner Summit). Snowfall in this region can significantly impact motorist traveling through mountain passes. Every year the California Department of Transportation battles to keep the major passes open during snow storms. Many of the motorists are traveling to destination ski resorts that are scattered around the mountain passes, while others support vital commercial shipping. These are only some of the users that rely on, and expect, accurate snowfall forecasts for this region.

The majority of the significant snowfalls across the Sierra Nevada are the result of Pacific weather systems that are steered into northern California by the powerful eastern Pacific polar jet stream. Many of these systems absorb moisture from maritime subtropical regions. Weaver (1962) classified these storms based on the longitude position of the downstream upper-level Pacific ridge (blocking high) which governs the latitude at which the surface lows approach California. The systems discussed in this paper were most similar to the high latitude type of the hydrological classifications that Weaver developed.

The dynamics and strong wind associated with a typical storm can produce several feet of snow. The most efficient precipitation producers appear to be the weather systems that close off or become strongly negatively tilted after the main cold front has been carried into the Sierra Nevada by the Pacific jet stream. This is because cold fronts will be accompanied by southwest to west wind in the 850 to 700-mb (approximately 5,000-10,000 ft MSL) layer at speeds of 25 to 65 kt (12 to 30 ms–1). The general northwest to southeast orientation of the Sierra Nevada, normal to this mean wind flow, allows for ideal orographic effects (i.e., upslope wind) that greatly enhance the precipitation rates. Since a closed or negatively tilted system becomes slow moving, additional precipitation occurs over the Sierra Nevada though the intensity is sometimes less because of weaker orographics. However, this can be countered with increased instability as the upper-level low pressure area advects colder air into northern California.

The trajectory of the Pacific polar jet stream during the winter months (storm track between 35 and 40° N) usually allows for subtropical moisture (high precipitable water values) to be advected northward from the lower latitudes (south of 30° N). The trajectory also allows for a long over- water fetch which entrains moisture well into the mid-tropospheric levels of the storm. On some occasions, subtropical moisture is advected directly over California as the subtropical jet stream shifts northward over the State. In this case, the Pacific polar jet will have phased with the subtropical jet stream due to the trough digging far south between 150 and 140° W. These types of synoptic patterns are often labeled "Pineapple Connections" because the moisture originates deep in the subtropics or the northern fringes of the tropics near Hawaii. Some of the large snow storms have been attributed to this type of synoptic pattern.

During the 2000 to 2001 season, the weather pattern over northern California was dominated by maritime polar air influxes. This study will focus on two significant snowfall events that occurred during the winter of 2001 that did not have a subtropical moisture source. The synoptic pattern during these events was dominated by a mid-tropospheric low pressure area over the Gulf of Alaska with a moisture source consisting of maritime polar air. An upper-level ridge of high pressure was positioned over the central Pacific which allowed several cold Gulf of Alaska low pressure areas to descend along the British Columbia coast and directly into northern California. This synoptic pattern still allowed for a connection to moisture from the northern Pacific (maritime polar) and a sufficient over-water fetch to maximize moisture entrainment (i.e., evaporation of ocean spray). During the event on 9–11 February 2001, two consecutive weather systems from the Gulf of Alaska deposited 48 to 68 in (122 to172 cm) of snow over the Sierra Nevada. These cold weather systems brought accumulating snowfall down to elevations of 1,500 ft MSL.

a. 7 April 2001

A late season snow storm on 7 April 2001 deposited 25 to 40 in (64 to 102 cm) of snow across the Sierra Nevada. Figure 1 shows the synoptic pattern that was observed during the event. A cold system moved across northern California with 500-mb geopotential heights lowering to below 540 dm. Snowfall rates during the peak of the storm on the early morning hours on 7 April were 3 in (7.5 cm) per hour. The GOES-10 water vapor imagery showed a well defined mid-tropospheric area of low pressure along the coast of California (Fig. 2). Precipitable water values on the Oakland (KOAK) sounding were observed to be 0.66 in (17 mm) at 0000 UTC 7 April. It is reasonable to suspect that the precipitable water values available for this system while it was over the Pacific Ocean could have originally been higher than that which was observed on the KOAK sounding due to vertical aliasing (the radiosonde would be carried inland). However, the water vapor image in Figure 2 does not show a subtropical moisture connection, and GOES precipitable water products did not show values over the eastern Pacific Ocean that exceeded 0.75 in (not shown).

b. 9-11 February 2001

Consecutive cold upper low pressure systems moved through northern California between 9 and 11 February 2001, leaving 48 to 68 in (122 to 172 cm) snowfall totals in the Sierra Nevada. Figure 3 shows a 500-mb geopotential height analysis which depicts the first upper-level low pressure area sliding along the British Columbia coast on 8 February. By 1200 UTC 9 February, this system had moved into northern California (Fig. 4) and snowfall rates of 2 to 3 in (5 to 7.5 cm) per hour were occurring in the northern Sierra Nevada. There was a short break in the heavy snowfall rates before the next upper-level low moved into the region. Figure 5 shows that the second system of similar strength moved onshore by 1200 UTC 12 February. Both of these mid- tropospheric circulations originated from the Gulf of Alaska and took very similar tracks. Geopotential heights dropped to slightly less than 540 dm across northern California. The KOAK sounding at 0000 UTC 10 February observed a precipitable water value of only 0.53 in (14 mm). As previously stated, this value may not be representative to what the upper lows originally experienced over the open water, but was considered acceptable based on satellite imagery.

a. Numerical Models

This study will not focus on numerical model performance during these events, however, it must be noted that models' forecasts of intensity and positioning of the mid-tropospheric low pressure systems were very accurate and consistent. An example of a 120-h forecast was included to show that even medium range model forecasts precisely depicted the synoptic pattern (Fig. 6). Comparing Figure 6 with Figure 5 shows how remarkably accurate was a 120-h forecast by the British Meteorology Office at Bracknell United Kingdom model (UKMET). Other numerical model forecast had similar performances. Figures 7 and 8 are BUFKIT (Mahoney and Niziol 1997) time-height sections of Eta model data at Blue Canyon (elevation 5,280 ft MSL) for both events. Blue Canyon's annual precipitation is near 68 in (1,725 mm) which is a representative average for much of the northern Sierra Nevada. Notice the high relative humidity and southwest wind forecast between 850 and 500 mb (see Figs. 7 and 8). For the 7 April event, the 1200 UTC 6 April Eta run underforecast precipitation at Blue Canyon since 2.19 in (55.6 mm) was observed compared to the 1.66 in (42.2 mm) forecast by the Eta (Fig. 7). During the period 9-11 February, Blue Canyon received 3.76 in (95.5 mm) of precipitation which was almost exactly the same total as the Eta 60-h forecast from the 1200 UTC 9 February run (Fig. 8).

Because there was not a subtropical moisture feed in these events, recognizing the dynamics, instability and orographics that were associated with these systems was the key to accurately predicting significant snowfall amounts. The mid-tropospheric low pressure systems originated from the Gulf of Alaska in a maritime polar air mass, but still had significant over-water fetches to obtain sufficient moisture levels. This is primarily a result of a strong low-level jet effectively entraining moisture from evaporation of ocean spray. The low-level jet and instability are subsequently due to latent heat of condensation release as this moisture condenses into the low- level cloud. These events showed the importance of synoptic-scale vertical ascent (omega) that results from strong differential positive vorticity advection in the mid troposphere combined with low to mid-level warm air advection. Despite precipitable water values under 0.75 in (19 mm), these strong dynamic processes maximized the condensation of much of the available moisture and enhanced precipitation rates resulting in heavy snowfall.

b. Cloud Microphysics and Orographics for Heavy Snow

It is also very important to consider the thermodynamic profiles that were present during these events. Both events showed that the saturated layer of the low- to mid-levels was located in temperatures between -10 and -18 °C. When strong vertical velocity is present in this mid-cloud layer and temperatures are between -12 and -18° C, snowflake production is most effective. At these temperatures clouds are more likely to contain ice nuclei. Therefore, the environment becomes increasingly supersaturated with respect to ice which makes the dentritic formation process most efficient. Since there was sufficient vertical velocity (pressure changes) at the proper levels during these events, snowflake generation and growth due to deposition and accretion were maximized (Auer and White 1982). In addition, even larger snowflakes can be produced in the warmer clouds (low cloud layers near 0 °C) that were present over lower elevations since aggregation is maximized near this temperature (sticking of snowflakes). Therefore, the efficiency of the cloud microphysics produced the most snowflakes and greater snowfall rates. Figure 8 shows how the forecaster can obtain the snow growth potential information from BUFKIT, and possibly use this to adjust their snowfall forecasts if other conditions are met.

The low-level forcing over the Sierra Nevada during these events was strong; however, it was not directly related to a deepening low-level cyclone (i.e., increased pressure gradient force) as analyses did not show any further pressure falls within the low center (not shown). The strong wind flow over the terrain was rather the result of prior mid-tropospheric cyclogenesis that had produced a low-level jet maximum. In addition, the mid-tropospheric low actually weakened (filled) slightly by the time it reached the Sierra Nevada. During each event, the low-level cyclone moved through Oregon and Nevada while it pushed a trailing cold front through northern California. The northwesterly to southeasterly orientation of the Sierra Nevada provides significant additional atmospheric lift as the air flowing from a near-perpendicular angle is forced to ascend upward along the mountain range. This enhanced vertical motion leads to the most efficient production of snowflakes. During the events in this study, and typical to many storms, southwesterly winds of 10 to 25 ms-1 (20 to 50 kt) were observed in the 850 to 700-mb layer (see Fig. 1). This resulted in strong orographic lift and precipitation rate enhancement.

In addition, the cold nature of these weather systems needs to be considered. A milder subtropical or modified subtropical air mass in a typical Sierra Nevada snow storm produces observed average snow to liquid ratios of 6 to 8:1 inches. The polar maritime events in this study had snow to liquid ratios on the order of 10 to 12:1 inches (not shown). Therefore, the lower water content of the snow (drier snowflakes) resulted in deeper total snow accumulations (less settling) over the higher elevations than compared to the wetter snowfalls.

Conclusion

The results of this study show that significant snowfall can occur when upper-level low pressure systems have a maritime polar origin. In order to accurately forecast snowfall amounts in the Sierra Nevada, the forecaster must consider the dynamics, instability and orographics associated with these systems rather than focus on available moisture sources. An understanding of cloud microphysics also needs to be factored into the snowfall forecasts since favorable cloud temperatures combined with sufficient moisture and vertical velocities can significantly increase snowfall rates. It has been shown that similar snowfall rates and totals, as those observed with the more typical Pacific weather systems, can occur with the colder storms that have no subtropical moisture source and originate from far northern latitudes near the Gulf of Alaska.

Acknowledgment

The author would like to thank Scott Cunningham (SOO at WFO Sacramento) for his review of this paper.

References

Auer, A. H. Jr., and J. M White, 1982: The Combined Role of Kinematics, Thermodynamics, and Cloud Physics Associated with Heavy Snowfall Episodes, J. Meteor. Soc. Japan, 60, 591-597.

Climate Prediction Center, cited 2001: [Available on-line from http://www.cpc.ncep.noaa.gov/products/intraseasonal/intraseasonal_faq.html]

Mahoney, E. A., cited 2000: BUFKIT Documentation. [Available on-line from http://www.nws.noaa.gov/er/buf/bufkit/bufkitdocs.html.]

Mahoney, E. A., and T. A. Niziol, 1997: BUFKIT: A software application toolkit for predicting lake-effect snow. Preprints 13th Intl. Conf. on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Long Beach, CA, Amer. Meteor. Soc., 388-391.

McLaughlin, M., 2001: The Reign of the Sierra Storm King, A Weather History of Donner Pass, California. Weatherwise, Jan.-Feb. 2001, 22-29.

Weaver, R. L., 1962: Meteorology of Hydrologically Critical Storms in California. U.S. Weather Bureau, Hydrologic Services Division, Hydrol. Rep. No.37, Washington D.C., 207 pp