JABUARY1959 MONTHLY WEATHER REVIEW 33

THE WEATHER AND CIRCULATION OF A MONTH OF EXCEPTIONAL PERSISTENCE FROM THEPRECEDING DECEMBER

L. P. STARK

Extended Forecast Section, U.S. Weather Bureau. Washington, D.C.

1. PERSISTENCE class changes(no change or achange of one class) for FROM DECEMBER 1958 TO JANUARY 1959 December toJanuary was71 percent. By comparison, WEATHER ANOMALIES January 1959 with 96 percent was phenomenally static. Fairly high persistence has been found in other Januarys Very high persistence over North America from Decem- (e.g., 1956 [2]), but in the last17 years there has not been ber to January was one of the outstanding facets of the a January withsmaller temperature changes than this monthly mean weather and circulation of January 1959. year. Parenthetically, it should be noted that 0+1 class One should anticipate anormal increase in month-to- temperature persistence greater than 90 percent has been month persistence of temperature, mean 700-mb. heights, observed only 17 times since April 1942 (8 percent of the and to a lesser degree, precipitation in passing from fall time). into winter [I]. However, stability of pattern and weather Precipitation from December 1958 to January1959 was anomalies from December to January was so great this also quite persistent. Table 1 showsa 47 percent value year that special emphasis is warranted. of the 0 class change in precipitation for January. This Class changes of monthly mean temperaturein the represents rather high stability whencompared tothe United Statesare shown intable 1. These data were December-Januaryaverage of 37 percentfor the years derived from an array of 100 well-distributed stations and 1942-1954 [I]. Infact, there were onlythree previous represent changes in classes based on the normal distribu- Januarys in the past 17 years in which the 0 class change tion of temperature at each city. The classes much above of precipitation was greater than the 47 percent of 1959. and much below normally occur 12:h percent of the time It is difficult to make direct comparisonof persistence of each, andabove, normal, and below normallyoccur 25 precipitationwith that of temperature since valuesare percent of the timeeach. On examining thetable, one computed for 0 class change for the former and 0+1 class is first struck by the absence of any change exceeding two change for the latter. However, the two can be compared classes. But evenmore significant, andcertainly more by expressing them in terms of their long-period average. indicative of the month's persistence, is the fact that 96 For example, this month's persistence of precipitation was percent of the stations did not change by more than one 127 percent of the 13-year average value, while the cor- class. responding figure for temperature was 135 percent. It is interesting to compare the results in table 1 with the averagesdetermined by Namias[I] for the period CIRCULATION 1942-1954. In his studythe averagenumber of O+l Persistence of the 700-mb. pattern was as pronounced as that of theweather anomalies.A qualitative com- TABLE1.-Class changes of weather anomalies in the United States parison of the 30-day mean 700-mb. chart for January from December 1958 to January 1959 1959(fig. 1) with that of December 1958 (fig. 1 of [3]) shows a ridge in western North America, a trough along the east coast, and a Low in Canada. Height departures Temperature fromnormal inthe same figures show a remarkable I similarity. A quant,itativeevaluation of circulation persistence can be expressed in terms of a lag correlation between the patterns of height anomalies. This was done earlier [l] Precipitation for the period 1933-1950, using a grid extending from 30' to 50O N. and from 70' to 130' W. An average lag cor- relation of +0.36 was obtained for all December to Janu- - ary periods. A similar correlation computed for Decem-

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC 34 REVIEW WEATHER MONTHLY JANUARY1959

FIQURE1.-Mean 700-mb. contours (solid) and height departures from normal (dotted) (both in tens of feet) for January 1959. Heavy dashed arrows represent the axis of the mean 700-mb. wind maximum. Widespread positive height departures over the polar region reflected intense blocking there. ber 1958 to January 1959 was +0.61, substantially greater sea level index in temperate latitudes averaged 1.6 m.p.s. than the average for the 18-year period above. below normal, following a near rxormal index in December. There was one feature of the circulation not wholly The configuration of systems typical of low index can be consistentwith thehigh persistence. A markedchange seen on the 30-day mean sea level chart (fig. 2). These occurred in the strength of the 700-mb. temperate wester- features include split Aleutian and IcelandicLows, strong lies (35'-55' N., computed twice monthly for overlapping polar Highs, and weak subtropical Highs. 30-day periods). The zonal index for the western half of the Northern Hemisphere forDecember was 12 m.p.s., 2. MAJOR FEATURES 0.7 m.p.s. above normal. During January the index de- OF THENORTHERN HEMISPHERE CIRCULATION creased to 9.9 m.p.s., l .9 m.p.s. below normal. This is the first month in the last nine in which the temperate index The hemisphericcirculation in January 1959(fig. 1) averaged below normal. The value of the30-day mean consisted of a wave number of four in middle latitudes.

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC JAHTJABY1959 REVIEWWEATHER MONTHLY 35

The troughs at higher latitudes in the western half of the Northern Hemisphere were rather weak compared with $hoseof the eastern half. This is shown by the relative intensity of the 700-mb. 30-day mean height departures from normal. The troughs near North America were less intense anddid not penetrate as far northward as did those of Europe and Asia. The circulation in midtroposphere in far northern lati- tudes of the Western Hemisphere was much more anti- cyclonic thanthat which normallyprevails [4]. The large High over the Beaufort Sea was an especially sig- nificant center of action in influencing weatherin the United States. It was part of a very extensive band of positive height anomalies which blanketed the area from the Bering Sea to Denmark Strait with heights ranging from 400 to 700 feet above normal. The unusual circula- tion in that region, which included anomalous centers of FIGURE2.-Mean sea level isobars (in millibars) for January 1959. the greatestabsolute magnitude in figure 1 and which Heavy dashedarrows representthe main anticyclonetrack in dominated the flow inthe Northern Hemisphere, was Januarv. The rsrominent ridge- from the Arctic Ocean to the Gulf also conspicuous in the weather and circulation regime of of M.exico shows the southward drive of cold air. ' December. Infact, high-latitude blocking over North America has been a circulation feature of varying magni- tude in each month since June 1958. A fairly well-developed Low in the Gulf of and its attendanttrough south-southwestwardmaintained the approximate intensity observed in December (fig. 1 of [3]), butimportant changes occurred. The troughac-

~ quired apositive tiltas the lower portionretrograded some 25' from its Decemberposition. As thetrough along the coast of Asia deepened, the wavelengt'h in the Pacific shortened. Thefast westerlies observedin mid- :;Pacific in December were replaced by a difluent zone near ' the International Date Line, and the mean 700-mb. wind speed maximum (shown by a dashed arrow in fig. 1) was displaced 10' southward in January from its posit'ion near Ihdd*- ""3 01 &bo". &xd on p.*Lm-w fclqCrPhr u s weatbn h" re@. -4. w, 37' N. in December (see fig. 3A of [3]). I FIGURE3.-Departure of average temperature from normal (" F.),1 Though the ridge-trough couplet over the United States January 1959. Hatchingindicates areas of normal or above and Canada appears to have been of small amplitude, the normal temperatures. The barrier effect of the Rockies is well flow was vigorous enough to keep the West warm and the shown as cold air stayed east of Divide. : rest of the United States quite cold (fig. 3). The 30-day mean 700-1000-mb. thickness anomaly field (fig. 4) shows (fig. I), accommodated the long wavelength downstream : the extent of the cold air and its extremely cold pool over tothe next trough in eastern Asia. Adequate cold air c northern British Columbia. The channel of cold air from supported the strong, full-latitude troughobserved in that northwestern Canada to Tennessee was practically coin- area, which is also a zone of high trough frequency clima- cident with a strong ridge of high pressure at sea level tologically [6]. (fig. 2) and with the path of maximum ant,icyclonic activ- ity, illustrated in figure 2 as a heavy dashed arrow. (See 3. VARIABILITY IN THE WESTERN HEMISPHERE also tracks of centers of anticyclones at sea level,Chart IX, DURING JANUARY I53. ) The ridge over Greenland, principally a development of Since the 30-day mean smooths shorter-term v~ri&o~ the first half of the month,was closely associated with the in the pattern, a breakdown of the mean into15-day com- strength of the trough in Europe. The westerly flow into ponents is of some interest. It has been noted above that central Europe met the northwesterlyflow from the Arctic the January patternclosely resembled that of the preced- in a zone of strong confluence, and cold air was thus con- ing month. Onemight thereby reasonably expect per- tained north of 50' N. in the mean (see fig. 4). sistencewithin themonth. But acomparison of the Strong westerlies across northern Asia, with the axis of circulation of the first and last halvesof the month (figs. 5 the mean 700-mb. wind speed maximum reaching 60" N. and 6) reveals rather sudden changes.

i

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC 36 MONTHLY WEATHER REVIEW JANUARY1959

FIGURE4.-Mean thickness (1000-700, mb.) departures from normal FIGURE5.-Mean 700-mb. contours (solid) and height departures (tens of feet) for January 1959. Below normal thickness areas fromnormal (dotted) (both in tens of feet)for -15, (shaded) were almost identical to areas of below normal surface 1959. Fast westerlies and a long half wavelength dominated the temperature (fig. 3). United States.

A downstream transfer of westerly momentum from the seen in figure 8 A and B, appear very much alike in spite coastal trough near Asia assisted in the erosion of the sub- of differences in circulation.However, an examination tropical ridge in themid-Pacific, where heightsfell as much of figure 8C shows significant changes related to the evolv- as 1000 feet (fig. 7). This allowed retrogression of the ing circulation. Note especially thenegative anomalous trough in the Gulf of Alaska (fig. 6), where heights in- temperature change in figure 8C and compare it with the creased over 300 feet, and its consolidation with the Kona heightchange in figure 7. Positiveheight anomalies of trough west of . January 1-15 were replaced by negativeanomalies of The moderately long wavelength between the troughin -30 from Hudson Bay to the Gulf of Mexico the eastern Pacific and that in the western Atlantic was andfrom the Continental Divide to the Ohio Valley. supported during the first half of January by rather strongAt t'hesame time, temperature anomaliesbecame more westerlies. But the half-wavelength downstream from the negative in the Plains and Central States as far south as ridge along the west coast was considerably longer than .Arkansas. Maximumchanges of10' F. were reported that usually observed, a condition which usually precedes nearthe Canadian border. Partially as a result of the a circulation change of major proportions. The vigorous new trough development, temperatures increased in the new trough over the Mississippi Valley in the succeeding South, the Northwest, and the East. An increase of 6' F. half-month (fig. 6) is an excellent example of a new mean occurred in Virginia in advance of the trough. trough development as described by Namias [7] and ap- The decreases of temperature in the Far West aremore parently was associated with: puzzling. An increase in 700-mb. height anomaly (fig. 7) a. A wavelength too long to be supported by the speed does not usually imply a decrease in temperature. Many of the westerlies. factors could contribute to a cooling, but perhaps most b. Retrogression of the ridge in western North Amerioa importantin this case was the direction of anomalous and its intensification as itamalgamated with the Arctic flow, which changed from southwest the first half of the Basin High. mont'h to northwest the last 15 days, as the ridge retro- c. Strengthening of the trough in central Canada. gradedfrom the Rocky Mountains to the west coast. d.Northeastward extension of the trough in Mexico. However,even though there was considerable intra- This classical development of a new trough is also a monthly cooling, meantemperatures were still above prime example of amplification of thecontour pattern. normalin the Far Westand Southwest for the period The intensity of the change from figure 5 to figure 6 is January 16-30. depicted in figure 7, the change in height anomaly from the first to the lasthalf of January. Meanwhile the trough 4. WEATHER IN THE UNITED STATES in the western Atlantic was swept eastward to the Azores, RELATED TO CIRCULATION where heights decreased over 400 feet.South of New- foundland height increases exceeded 700 feet as a strong TEMPERATURE ridge replaced the trough. Several injections of 'Arctic air into the United States Temperature anomalies for each half of the mont'h, as east of the Divide produced a cold regime in the East.

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC -.JANUABY 1959 MONTHLYREVIEW WEATHER 37

FIGURE6.-Mean 700-mb. contours (solid) and height departures from normal (dotted) (both in tens of feet) for January 1G30, 1959. Note new trough which formed in central United States as the ridge inthe West retrogradedand the trough in the western Atlantic progressed.

FIGURE7.-Change of mean 700-mh. height departure from normal (tens of feet) from January 1-15 to January 16-30, 1959. This pattern accompanied considerable amplificationof the long waves.

This. togetherwith warming in the West, resulted in a ' I IU - I distribution of temperature which was meridional,a FIGURE8,"Temperature departure from normal (" F.) for (A) situation generallyassociated with greattemperature January 1-15, 1959, (B) January 16-31, 1959. Shadedareas are extremes. In thismonth, forexample, thewarmest normal or above. In (A) warming east of Divide in Plains States January of record was observed at and San was related to foehn action. In (B) note cooling in North Central and Northern Plains States. (C) Change in temperature depar- Francisco, while Muskegon,Mich. and International ture from normal (" F.) from January 1-15 to January 16-31, Falls, Minn. were havingthe coldest (and snowiest at 1959. Shadedareas represent positivechanges (warming). Muskegon) January of all time.' (From 181.) Extreme cold earlyin January accompanied a front which dropped daily temperatures as much as50' and was below normal at Amarillo,Tex. [8]. As cold air spread responsible for a weekly averagetemperature of 20' to the east coast, warmingoccurred throughout the West.

1 From Local Climatological Data report of each city, January 1959. An example of the magnitude of tha.t change was recorded

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC 38 MONTHLYREVIEW WEATHER JANUABY1969

a

FIGURE9.-Percentage of normal precipitation for January 1959. Heaviest amounts were foundnear the mean 700-mb. wind maximum and cyclone track. (From [8].) FIGUREIO.-Mean sea level pressure departure from normal(milli- bars) and principal cyclone track (heavy dashed arrows). There was a close correspondence between the position of thestorm track,the mean 700-mb. windmaximum, andarea of heavy at Great Falls, Mont. where a minimum temperature of precipitation. -29O on was followed by a maximum of62’ on the 12th.‘ By midmonth, warming progressed to the Plains States, while the Eastern States had become quite cold. During the week ending , cold air again PRECIPITATION plunged to the Southern Plains States, following a fast- Total precipitation accumulation in the United States, moving storm which became well organized in New in terms of percentage of monthly normal, are shown in Mexico on . Average temperaturesin Okla- figure 9. The mostprominent feature is the band of homa, which were 6’ a,Eove normal the preceding week, abovenormal precipihtion which extendedfrom the dropped to 6’ below normal,and Birmingham, Ala. Pacific Northwest (where Spokane had its wettest Janu- recorded a 52’ decrease in temperat’ure in 18 hourson the ary), across Kansas,through the Ohio Valley, and into 21st. The reversechange intemperature was reported New .Disastrous flooding occurred in Ohio, in the East from the Carolinas northward and resulted western New York (where Buffalo had its wettest Janu- in a spectacular January thaw in the manner described ary), western , and along the Wabash River by Wahl [9]. systemin Indiana. Rainfall of recordproportions and January was not especially noteworthyregarding melting snow associated with the storm of January 20-23 temperature records, but there were additionalfeatures were responsible for the flood loss of many lives and of interest. A comparison of figures 3 and 4 shows that property loss of millions of dollars. the isopleth of 0’ temperature departure from normal is It is generally harderto relate precipitation to the almost identicalwith the zero isopleth of theaverage observed circulation or its anomalous flow than temper- thickness DN. One also sees that below normalthick- ature. In thisJanuary, however, thetask isrelatively nesses almost everywhere match below normaltemper- simple. In figure IO, the 30-day mean sea level pressure atures. It should also be noted that the height anomalies departures,one can see a corridor of negativevalues of figure 1 have a shnilar relationship t,o the temperature, which closely resembles the configuration of thearea although less well defined. of heavy precipitation. This was related to the principal The 30-day mean of sea level pressure (fig. 2) also st,orm trackduring January, represented by the heavy helps explain thetemp.erature distribution of figure 3. dashed arrow. The storm track is based on an analysis The ridge from the Beaufort Sea to Florida and the flow of frequencies of cyclones in a grid of equal. area..”xes from theDistrict of Mackenzie tothe Atlantic Coast and on inspection of tracks of daily centers of cyclones States defines the source of cold airand its transport at sea level (Chart X of [5]). Note also theproximity southeastward. Thewarm High over the Great Basin of the 700-mb. mean wind speedmaximum (in fig. 1) and the onshore flow into the West are indicative of the tothe cyclone trackand axis of hwy pmcipitation. above normal temperatures observed there. It is of some Compared with normal tracks and mean frequencies of historical significance that coastal areas have cyclones [lo], those in January were displaced southward escaped the below normalcategory since early April 5’-10’ of latitude, presumably as a consequence of high- 1958, except for twoconsecutive 5-day periods inmid- latitude blocking andthe resultant depression of the November. westerlies.

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC JANUARY 1959 MONTHLY WEATHER REVIEW

Referring again to figure 9, other areas of above normal 3. R. A. Green, “TheWeather and Circulation of December precipitation are shown in California, where Mt. Shasta 1958,” Monthly WeatherReview, vol. 86, NO. 12, Dec. 1958, pp. 487-492. had the wettest Januarysince 1916. There was a second- 4. U.S. WeatherBureau, “Normal Weather Charts for the ary streak from northern Texas to South Carolina. The Northern Hemisphere,” Technical Paper No. 21, 1952, 74 pp. large area which had less than 60 percent of normal 5. U.S. Weather Bureau, Climatological Data, National Summary, precipitation included Dallas,Tex., which reportedthe vol. 10, No. 1, Jan. 1959. driest January of record,and Albuquerque, N. Mex. 6. W. H. Klein and J. S. Winston,“Geographical Frequency of Troughs and Ridgeson Mean 700-mb. Charts,” Monthly which reported its sunniest. Dry weatherin this area Weather Review, vol. 86, No. 9, Sept. 1958, pp. 344-358. was related to northwesterly flow aloftsouth of the 7. J. Namias, Extended Forecasting by Mean Circulation Methods, principal storm track. U.S. Weather Bureau, Feb. 1947, 89 pp. 8. U.S. Weather Bureau, Weekly Weather andCrop Bulletin, Na- REFERENCES tional Summary, vol. XLVI, No. 1, Jan. 5, 1959, and NO. 8# 1. J. Namias, “The Annual Course of Month-to-Month Persistence Feb. 23, 1959. in Climatic Anomalies,” Bulletin of the American Meteoro- 9. E. Wahl, “The January Thaw in New England,’’ Bulletin of the logical Society, vol. 33, No. 7, Sept, 1952, pp. 279-285, and an American Meteorological Society, vol. 33, No. 9, NOV.1952, unpublished extension through 1954. pp. 380-386. 2. W. H. Klein, “The Weather andCirculation of -A 10. W. H. Klein, “PrincipalTracks and Mean Frequencies of Month with a Record Low Index,” Monthly Weather Review, Cyclones and Anticyclones inthe Northern Hemisphere,’’ vol. 84, No. 1, Jan. 1956, pp. 25-34. U.S. Weather Bureau, Research Paper No. 40, 1957, 40 pp.

Unauthenticated | Downloaded 09/26/21 10:02 PM UTC