FEBRUARY, 1990. MONTHLY WEATHER REVIEW. 73
times when the coiiditiou is well-dc~-eloped R fine ri:iii fdls covering everything with R coitt of glaze ice.
At such times nu esamiiiation of the weather map shows we are to the north of the nsis of a trough of low prcssure xiid thnt but n little wily south of us wnrni west to southwest minds are I)lowii?g wid 110 douht over-
riding this cold northerly air, producing the clouds ruirl
sometimes rtiin. As a rule t,his coiditioii is followcd !I)-
ik shift of wind to SW. and :t riipiel rise in temper:it.urc, the clouds all disappenriiig.
BOUNDARY BETWEEN A SOUTH WIND AND AN UNDER- RUNNING NORTHEAST WIND.'
By CHARLES F. HROOKN.
[Blue Eill Observatory, Mass., Slay, 1913.1
In t.he 1nt.c nfternoon and o.vc:riitig of AIilX 3, 1C113, :LE
finticyclone began to approach Bluc Hill from t.ho wirth. Duririg the day n soa.-lwecze of three hours' durntion had preveiitcd the masimum temperaturc (80°F.) from iiccurrin until the middle 41f the nftcrnooii at, tho sunimit of the hfrl (195 m.'i. The wind roiiinitiucl i i i ai southcrrl-
direction till 9 11. m. with momen t,il.ry esccptiotis, mwkd hy, sharp drops in the therinograph cur\-c.
f i c wind at the base station (64 m.) heciune iiortlioi:st,
at 5:2O p. m. nnd therefore for sererid !lours il consitler-
of wmrwhat simlln; rtmdition?.
- - -- - . __ -. - . . . . - . . . - .. . - . . .. . . . . .
1 Sfc C. F. Brooks Thee ice storma, Srimrt, Auj?. S, 1Y1.1, pp. 12bIY-I. for drwrI]~riclna
iihla tenipcrii ture iuversion esisted between the I>me
:i t i d summit, tit one time nmouiibing to 14'F.
the dividing liue hetween the iiorthenst witic?a:f.gi south-wuthe:rst wild was across the summit of the hill
iri n.11 cast-west cliri?ctioii. This liiic was very shnrp, so-
sh::rp that I could stnud with the 1)reeze from the north- mist hlowitig 011 wit- cheek atid that froin the south 011
bho other. At, this time, os shown I)y the osc.illatioiis of thc? hygrogr:ipb curve up :i.ncl h c k ngniri wlieiiever II
siii:ill voluiii~~ of il(&leilst. wind ldew through tlie instru- metit slielt.er, tlic reliitive humidity o!i the dividiirg line
wiis Yo per cent oii otic side (northenst') :i d 40 per ceiit
oil the othe.r (sciuth!. With t,he normal ewiiit!g cooliiig thc! humidity pres-
is!itly re:ichr.d 100 per c.eiit nil the dividing zone, thereijy l~roduci~ig I! fo . The pressure hegn.:i its rapid rise a t this time. Bekrc long the stratus cloud rose from the top of the hill (3 :I. m.? hy hj-grogrnph). and liy 7 a. in.,
Mn.1- 4, wiis soiiic diutm:re e h v e , its appe:irmice giriiig
i i i i iiircrsc. .'ninn-mtito" effect. At rrhout, noon the suii
first hrolx through the chud sheet, nncl hy sundown most was go11e.
(NOTE. --hother occasion on which I observed n. simi- 1:tr s1i;q~ el ividii;g line between wiiids from two direc-
tioris w'us 01: July 35, 1912 near t.lie Schilthorn, Switzcr1::iid. III the late :jfterrtooii, wheii there were thuiidei~torins i :i tlie vicinity, n clnudy curreiit comiiig
up the south side of a ridge o :~ the eirst nict n clear
mrreil t 11 j t,he north sidv with t.lic result t h i t i\ \-ertic.?nl c h i d mi 1 I perhuys 100 nict.ers high WIS formed.)
THE PRECIPITATION OF SLEET AND THE FORMATION OF GLAZE IN THE EASTERN UNITED STATES, JANUARY 20 TO 25, 1920, WITH REMARKS ON FORECASTING.
SYNOMS.
h i attempt is iiide. b:; nieans of wcurst.c charts of precipitai.icm during the previous 12 hours, ciirreut. twoperatwe. ressurr, aiid linw
of \r i d flow. i n combination wit.h such nerolog4i.aPdat.a 3s rould he nhtained. to construct I.-~WY-SP&CBJIS r i l thr Inner 3 kilcmwtcw of t.hv
atiucsphero. during tho period January 20 to 35. 1920. From such charts arc? shown tho actiid procwscs which produce rain. rli~et,. and mow. ~eparately and in conitination. in swh n maimer a.s til prodiiw the ic-e ooyer. whivh is called an '.ice utorm." Thc rondition is. briefly. a cold northerlv wind underrunning a v:arvil sniith~rly c.iirrent. forcing the 1att.er doit. The vertival dirtriliuticw of tcwipergture. ahown in the crass-aectiona. inclivates the inanwr in w11ic.h thv i~i~tli~lrins in t.hat. territory covered hy the nort.herly wind rise norn~ally rintil t.hs level of tho overrunning southerly wind is at,tt.tniiwd. wirwe thr isot.lierni swrrws shnrplv north~-nrcI. 'rhc distaiirc! that thc iwthwni of freezing rewhm i R indicated by tho northern limit of the precipitn- tion of sleet. .\n r!mpirid reLt,ion WTUF obtain4 b;?tween t.hc distanrc knni the
wind-shift line to the 38' isothcrui mid (J ) t.he width of the glxrr! hdt, (8) t.he niclbh of the deet helt, 1.31 t.hr distance of the venter i?f
the sleet belt north of t.hr 33' isntherm, (4) t.hr? width of t,hP d n z e liclt on a meridian 4 O east 12 houm latcr. and i.i) the width of l.hr ylaxe belt on u meridian So ewt 24 hours lab. 'I'hrse ~a l u e ~ are presented \Fit.h the full realization that. they may Im true fnr this particuIa1 storm
U I I ~~, and are as foliowj: (1 ) The width of the glzlaxe bdt=thc distance between the :Eo isotherm and the wind-shift, line: (2) The width of the sleet. belt=O.'i x tho distance between t.he 22' isotherm and t,he wind-shift line;
(3) The distance between thr? Reo isotherm and t.he center of t h r sleet helt=0.8 x the rlisbnncv Let,v.een the RSoiwtherm and thewind- shift line;
(4) The width of t.he glaze belt 4' met, 18 houm lat.er=O.!I x the distance between the :Eo isotherin and the wind+hift line; and,
(5) The width of the glaze k~elt 8' east., 21 hours later=0.8 )( t,he diitance between the 52' imtherni and the wind-shift. line.
The importanw of Lhe wind-.shift line in lorerlssting the region oyer !vhivh slpet or *lsxo are likely to occur is stron7ly emphsixed. R i i i w
it mark* the p:int. of axwit. qf the rrouthr.rly wind and henr-e i4 the bwirr upon w1iic.h rwtr t.hrB lomtiw of this t.) pc nf prel.ipit.irtinn.
IXTRODUCTIOS.
Of all ty es of storms, there are fcw which have the wide-spreac r econoniic effects of the so-called "ic.e storm."
Not only is traffic, both 011 railroads uncl in citios, im- peded and often completely tied up, and telephone and telegraph lines crippled, but accidents are numerous also. Moreorer, when rain falls on x region previously covered with snow and produces a11 ice glaze, the snow is held immovable and the glaze forms x gliding surface over 1vhic.h subscquent. snow will drift with little hindrance.
I11 New York City recently, when tlie streets were more effectually blockaded by snow than at any time in the city's hist,ory, not a small part of the difficulty in its removal was att,ributnble to the fact that there had been layers of ice formed at various levels in the snow, in- creasing the rigidity of the drift.s and packing them more solidly. Often such storms are local and do not have a wide- spread effect., but once or twice in a winter they occur over a largo area of the country. They are caused, of course, simply hg the recipitation of ram upon a region
many cases, it is not long before all su aces esposed to the rain become heavily coated with a crystal-clear la er of ice, sometimes as much a8 an inch in thickness. d i s
3
the temperature of w g ich is a t freeziii or below. In
74 MONTHLY WEATHER REVIEW. FEBRUARY, 19‘20
type of storm has been called by the rather unsatisfactory name of ice s t m n . The ice cover which is formed is
called by the English gZazed rost; this is not satisfactory
the German Glatteis are more descriptive. Our word gluze, adopted by the Weather Bureau in 1916, comes nearest to a satisfactory designation for this condition.‘ The conditions under which such preci itation is pro-
Dr. C. F. Brooks,a and the reader is referred to his article for an explanation of the processes taking place aloft which produce areas of snow, sleet, glaze, and rain, either singly or ih combination. This paper proposes to discuss, as far as available data and legitimate specula- tion will permit, the conditions aloft as well as at the surface, and to show how they interact in a specific
storm.
because it is not related to f rost; the French vergZas and
duced have been discussed in the preceting 7 article hy
PLAN OF THE STUDY.
The winter of 1919-20 has been a particularly unpleas- ant one in the Emtern United States, owing, especially, to the eat precipitation of snow in the northern portions, and o !? snow, sleet, and rain in the middle latitudes, which formed a solid, slow-melting cover. Perhaps the worst eriods of this t pe were t>hose of January 20-25
territor east of the 100th meridian, while the lat8ter was confine i to the Atlantic coast States. The first period has been investigated the more thoroughly owing to its greater mea and duration.
and Fe \ n a r y 3-6,’the s ormer covering practically all the
which so
and glaze formation:
1. Low temperature and high pressure to the northward (between northwestandnortheast.) * * * 2. Steep pressure and temperature gradients to the nurthward (he-
tween northwest and northeast.) * * * 3. Surface temperatures below freezing. * * * 4. Moderately high pressure and high temperature over the East Gulf and South Atlantic States. * * * 5. Northward looping of the isotherms. * * * 6. Gentle to fresh northerly winds, increasing by time sleet begins. * * * 7. Low prmre trough trending southwest to northeast between two
HIGHS. The LOW is uwally moving from the southwest, but sometimes from the northwest.
And, while it is ossible, when such formations are pres-
is rarely ossible to forecast its location more definitely
is, obvious1 , a function of the temperature both at t8he surface angaloft, and of the strength of the north and south components of the underrunning and overriding winds. Even though the critical exmination of a single storm is unlike1 to shed as much li h t on these relation-
ation justifies itself in settin forth clearly throughout
ent, to say that a e eet d l occur or that glaze will occur, it
than to d!) eaignate its-center. But the width of the area
ships as might i e desired, neverthe K ess, such an investi-
%e enod the conditions whic E did prevail together with the Ln d of precipitation and the regions over which it occurred. Meridional sections o f the atmosphere.-Besides the ordi- nary relations of barometric pressure and precipitation, it -__~
1 Abbe. Cleveland. Jr.: American deflllltlon of “&at.” MONTHLY WEATBEB RE- I ~W igie, u: asi-~a: : Brooks Charlea F.: Thenature of sleet and how it Is formed Thls REVIEW pp. 69-72.
*U 8 be of A culture, Weather Bureau: Weather iorecsatinp in {he United W& $6, pp. %7-m.
is necessary to investigate by means of meridional cross- sections the distribution of temperature, cloudine,ss, and wind velocit,y aloft; also to construct maps showing the areas over which precipitation in the form of rain, sleet, or snow, or cnmbinations of these, occurred. I n con-
structing the meridionnl cross-sections two assumptions are nec.essary-first, that the wind of southerly component continues aloft after le,aving the ground, overriding the wind of nort,herly component; and, second, t>hat the tem- perature of the, wind of southerly component ma.y be sur- mised froni the following: (a1 The temperature fall per degree of latitude as t,he southerly wind procee.ds north-
ward can give a clue as to the temperature of the wind afte,r it has risen from the surface; (6) the dynamic cool- ing of the southerly air ns i t rises can be surmised from t,he amount of precipit.ation: (c ) the northern limit of the isotherm of fre.ezing is marked, wit,h due cnnsiderat,ion for the effect of wind velocit,y in keepin the sleet pellets
limit of sleet or rainfall; and (d) the motion, type, and height of the base of the clouds tell much concerning t,he conditions aloft, where kite. observations are’wanting. Width of sleet and gh.zi> belts.-The width of the sleet belt and glaze belt, are, of course, of prime importance in the, forec.ast, and thus the forecast becomes prinisrily one of temperature; i. e., hhe prechting as accurately as possible. the location of the horizontal freezing line. But while this can he n certain guide to the general loca- tion of sleet and glaze, there must be othex conditions peculiar to the winds involved whidi will determine the widt,h of the are,a over which preci itation will occur.
gested: (I) The steepness of the surfnce temperature gradient within the zone of northerly winds; (a) the st,rength of the southward component in the northerly wintls; (3) the strength of the southerly wind aloft; (4)
the difference. between tlie su1.fac.e temperature and the maximum temperature a,loft.; and (5 ) the determination
of tlic height of the maximum temperature aloft. The last point is one that is subject to several considerations.
I€ one considers the surface t)emperature in the southerly wind at some point south of the wind-shift line, or the line where tlie southerly com onent disappears and the
priat’e vertic.al temperature gradient, dependent upon the current conditions. Direct observation by kites is desirable, but where this is impossible, the current condi- tions can be surmised by allowing the vertical temper- ature gradient to depart from the normal shown by kites in accordance with the variation of ot1ie.r condit.ions from the normal, such as wind speed and snow cover. Again, after this southerly current le,aves the ground its temperature must be cl-jmamically lowered. Finall , the strength of the sout,herly component must be consi B ered, for it is ap arent that an increase of wind velocity with
high relative to that at the surface, is certain to carry the warm air farther north over the cold layer than would otherwise be the case. All of these three factors must be included in any speculation regarding the altitude of the freezing tem erature at its northern extre.mity aloft.
various standpoints mentioned above, the accompany- ing series of charts has been devised. The even-num- bered Charts 11, IV, VI, etc., consist of two maps, the upper showing the sea-level pressure distribution, and the lower showing the instantaneous stream lines at the same time. The stream line map must not be misunder-
and rttindrops from falling vertically, % y the northern
At the outset the following consir P erations were sug-
northerly ap enm, it is possib Y e to approximate the alti- tude of the P reezing temperature by assuming an appro-
alt,itude, w 7l en the wind is bearing a. temperature very
Cha.Tta.- +) o aid in the discussion of the data from the
74 MONTHLY WEATHER REVIEW. FEBRUARY, 19‘20
type of storm has been called by the rather unsatisfactory name of ice s t m n . The ice cover which is formed is
called by the English gZazed rost; this is not satisfactory
the German Glatteis are more descriptive. Our word gluze, adopted by the Weather Bureau in 1916, comes nearest to a satisfactory designation for this condition.‘ The conditions under which such preci itation is pro-
Dr. C. F. Brooks,a and the reader is referred to his article for an explanation of the processes taking place aloft which produce areas of snow, sleet, glaze, and rain, either singly or ih combination. This paper proposes to discuss, as far as available data and legitimate specula- tion will permit, the conditions aloft as well as at the surface, and to show how they interact in a specific
storm.
because it is not related to f rost; the French vergZas and
duced have been discussed in the preceting 7 article hy
PLAN OF THE STUDY.
The winter of 1919-20 has been a particularly unpleas- ant one in the Emtern United States, owing, especially, to the eat precipitation of snow in the northern portions, and o !? snow, sleet, and rain in the middle latitudes, which formed a solid, slow-melting cover. Perhaps the worst eriods of this t pe were t>hose of January 20-25
territor east of the 100th meridian, while the lat8ter was confine i to the Atlantic coast States. The first period has been investigated the more thoroughly owing to its greater mea and duration.
and Fe \ n a r y 3-6,’the s ormer covering practically all the
which so
and glaze formation:
1. Low temperature and high pressure to the northward (between northwestandnortheast.) * * * 2. Steep pressure and temperature gradients to the nurthward (he-
tween northwest and northeast.) * * * 3. Surface temperatures below freezing. * * * 4. Moderately high pressure and high temperature over the East Gulf and South Atlantic States. * * * 5. Northward looping of the isotherms. * * * 6. Gentle to fresh northerly winds, increasing by time sleet begins. * * * 7. Low prmre trough trending southwest to northeast between two
HIGHS. The LOW is uwally moving from the southwest, but sometimes from the northwest.
And, while it is ossible, when such formations are pres-
is rarely ossible to forecast its location more definitely
is, obvious1 , a function of the temperature both at t8he surface angaloft, and of the strength of the north and south components of the underrunning and overriding winds. Even though the critical exmination of a single storm is unlike1 to shed as much li h t on these relation-
ation justifies itself in settin forth clearly throughout
ent, to say that a e eet d l occur or that glaze will occur, it
than to d!) eaignate its-center. But the width of the area
ships as might i e desired, neverthe K ess, such an investi-
%e enod the conditions whic E did prevail together with the Ln d of precipitation and the regions over which it occurred. Meridional sections o f the atmosphere.-Besides the ordi- nary relations of barometric pressure and precipitation, it -__~
1 Abbe. Cleveland. Jr.: American deflllltlon of “&at.” MONTHLY WEATBEB RE- I ~W igie, u: asi-~a: : Brooks Charlea F.: Thenature of sleet and how it Is formed Thls REVIEW pp. 69-72.
*U 8 be of A culture, Weather Bureau: Weather iorecsatinp in {he United W& $6, pp. %7-m.
is necessary to investigate by means of meridional cross- sections the distribution of temperature, cloudine,ss, and wind velocit,y aloft; also to construct maps showing the areas over which precipitation in the form of rain, sleet, or snow, or cnmbinations of these, occurred. I n con-
structing the meridionnl cross-sections two assumptions are nec.essary-first, that the wind of southerly component continues aloft after le,aving the ground, overriding the wind of nort,herly component; and, second, t>hat the tem- perature of the, wind of southerly component ma.y be sur- mised froni the following: (a1 The temperature fall per degree of latitude as t,he southerly wind procee.ds north-
ward can give a clue as to the temperature of the wind afte,r it has risen from the surface; (6) the dynamic cool- ing of the southerly air ns i t rises can be surmised from t,he amount of precipit.ation: (c ) the northern limit of the isotherm of fre.ezing is marked, wit,h due cnnsiderat,ion for the effect of wind velocit,y in keepin the sleet pellets
limit of sleet or rainfall; and (d) the motion, type, and height of the base of the clouds tell much concerning t,he conditions aloft, where kite. observations are’wanting. Width of sleet and gh.zi> belts.-The width of the sleet belt and glaze belt, are, of course, of prime importance in the, forec.ast, and thus the forecast becomes prinisrily one of temperature; i. e., hhe prechting as accurately as possible. the location of the horizontal freezing line. But while this can he n certain guide to the general loca- tion of sleet and glaze, there must be othex conditions peculiar to the winds involved whidi will determine the widt,h of the are,a over which preci itation will occur.
gested: (I) The steepness of the surfnce temperature gradient within the zone of northerly winds; (a) the st,rength of the southward component in the northerly wintls; (3) the strength of the southerly wind aloft; (4)
the difference. between tlie su1.fac.e temperature and the maximum temperature a,loft.; and (5 ) the determination
of tlic height of the maximum temperature aloft. The last point is one that is subject to several considerations.
I€ one considers the surface t)emperature in the southerly wind at some point south of the wind-shift line, or the line where tlie southerly com onent disappears and the
priat’e vertic.al temperature gradient, dependent upon the current conditions. Direct observation by kites is desirable, but where this is impossible, the current condi- tions can be surmised by allowing the vertical temper- ature gradient to depart from the normal shown by kites in accordance with the variation of ot1ie.r condit.ions from the normal, such as wind speed and snow cover. Again, after this southerly current le,aves the ground its temperature must be cl-jmamically lowered. Finall , the strength of the sout,herly component must be consi B ered, for it is ap arent that an increase of wind velocity with
high relative to that at the surface, is certain to carry the warm air farther north over the cold layer than would otherwise be the case. All of these three factors must be included in any speculation regarding the altitude of the freezing tem erature at its northern extre.mity aloft.
various standpoints mentioned above, the accompany- ing series of charts has been devised. The even-num- bered Charts 11, IV, VI, etc., consist of two maps, the upper showing the sea-level pressure distribution, and the lower showing the instantaneous stream lines at the same time. The stream line map must not be misunder-
and rttindrops from falling vertically, % y the northern
At the outset the following consir P erations were sug-
northerly ap enm, it is possib Y e to approximate the alti- tude of the P reezing temperature by assuming an appro-
alt,itude, w 7l en the wind is bearing a. temperature very
Cha.Tta.- +) o aid in the discussion of the data from the
74 MONTHLY WEATHER REVIEW. FEBRUARY, 19‘20
type of storm has been called by the rather unsatisfactory name of ice s t m n . The ice cover which is formed is
called by the English gZazed rost; this is not satisfactory
the German Glatteis are more descriptive. Our word gluze, adopted by the Weather Bureau in 1916, comes nearest to a satisfactory designation for this condition.‘ The conditions under which such preci itation is pro-
Dr. C. F. Brooks,a and the reader is referred to his article for an explanation of the processes taking place aloft which produce areas of snow, sleet, glaze, and rain, either singly or ih combination. This paper proposes to discuss, as far as available data and legitimate specula- tion will permit, the conditions aloft as well as at the surface, and to show how they interact in a specific
storm.
because it is not related to f rost; the French vergZas and
duced have been discussed in the preceting 7 article hy
PLAN OF THE STUDY.
The winter of 1919-20 has been a particularly unpleas- ant one in the Emtern United States, owing, especially, to the eat precipitation of snow in the northern portions, and o !? snow, sleet, and rain in the middle latitudes, which formed a solid, slow-melting cover. Perhaps the worst eriods of this t pe were t>hose of January 20-25
territor east of the 100th meridian, while the lat8ter was confine i to the Atlantic coast States. The first period has been investigated the more thoroughly owing to its greater mea and duration.
and Fe \ n a r y 3-6,’the s ormer covering practically all the
which so
and glaze formation:
1. Low temperature and high pressure to the northward (between northwestandnortheast.) * * * 2. Steep pressure and temperature gradients to the nurthward (he-
tween northwest and northeast.) * * * 3. Surface temperatures below freezing. * * * 4. Moderately high pressure and high temperature over the East Gulf and South Atlantic States. * * * 5. Northward looping of the isotherms. * * * 6. Gentle to fresh northerly winds, increasing by time sleet begins. * * * 7. Low prmre trough trending southwest to northeast between two
HIGHS. The LOW is uwally moving from the southwest, but sometimes from the northwest.
And, while it is ossible, when such formations are pres-
is rarely ossible to forecast its location more definitely
is, obvious1 , a function of the temperature both at t8he surface angaloft, and of the strength of the north and south components of the underrunning and overriding winds. Even though the critical exmination of a single storm is unlike1 to shed as much li h t on these relation-
ation justifies itself in settin forth clearly throughout
ent, to say that a e eet d l occur or that glaze will occur, it
than to d!) eaignate its-center. But the width of the area
ships as might i e desired, neverthe K ess, such an investi-
%e enod the conditions whic E did prevail together with the Ln d of precipitation and the regions over which it occurred. Meridional sections o f the atmosphere.-Besides the ordi- nary relations of barometric pressure and precipitation, it -__~
1 Abbe. Cleveland. Jr.: American deflllltlon of “&at.” MONTHLY WEATBEB RE- I ~W igie, u: asi-~a: : Brooks Charlea F.: Thenature of sleet and how it Is formed Thls REVIEW pp. 69-72.
*U 8 be of A culture, Weather Bureau: Weather iorecsatinp in {he United W& $6, pp. %7-m.
is necessary to investigate by means of meridional cross- sections the distribution of temperature, cloudine,ss, and wind velocit,y aloft; also to construct maps showing the areas over which precipitation in the form of rain, sleet, or snow, or cnmbinations of these, occurred. I n con-
structing the meridionnl cross-sections two assumptions are nec.essary-first, that the wind of southerly component continues aloft after le,aving the ground, overriding the wind of nort,herly component; and, second, t>hat the tem- perature of the, wind of southerly component ma.y be sur- mised froni the following: (a1 The temperature fall per degree of latitude as t,he southerly wind procee.ds north-
ward can give a clue as to the temperature of the wind afte,r it has risen from the surface; (6) the dynamic cool- ing of the southerly air ns i t rises can be surmised from t,he amount of precipit.ation: (c ) the northern limit of the isotherm of fre.ezing is marked, wit,h due cnnsiderat,ion for the effect of wind velocit,y in keepin the sleet pellets
limit of sleet or rainfall; and (d) the motion, type, and height of the base of the clouds tell much concerning t,he conditions aloft, where kite. observations are’wanting. Width of sleet and gh.zi> belts.-The width of the sleet belt and glaze belt, are, of course, of prime importance in the, forec.ast, and thus the forecast becomes prinisrily one of temperature; i. e., hhe prechting as accurately as possible. the location of the horizontal freezing line. But while this can he n certain guide to the general loca- tion of sleet and glaze, there must be othex conditions peculiar to the winds involved whidi will determine the widt,h of the are,a over which preci itation will occur.
gested: (I) The steepness of the surfnce temperature gradient within the zone of northerly winds; (a) the st,rength of the southward component in the northerly wintls; (3) the strength of the southerly wind aloft; (4)
the difference. between tlie su1.fac.e temperature and the maximum temperature a,loft.; and (5 ) the determination
of tlic height of the maximum temperature aloft. The last point is one that is subject to several considerations.
I€ one considers the surface t)emperature in the southerly wind at some point south of the wind-shift line, or the line where tlie southerly com onent disappears and the
priat’e vertic.al temperature gradient, dependent upon the current conditions. Direct observation by kites is desirable, but where this is impossible, the current condi- tions can be surmised by allowing the vertical temper- ature gradient to depart from the normal shown by kites in accordance with the variation of ot1ie.r condit.ions from the normal, such as wind speed and snow cover. Again, after this southerly current le,aves the ground its temperature must be cl-jmamically lowered. Finall , the strength of the sout,herly component must be consi B ered, for it is ap arent that an increase of wind velocity with
high relative to that at the surface, is certain to carry the warm air farther north over the cold layer than would otherwise be the case. All of these three factors must be included in any speculation regarding the altitude of the freezing tem erature at its northern extre.mity aloft.
various standpoints mentioned above, the accompany- ing series of charts has been devised. The even-num- bered Charts 11, IV, VI, etc., consist of two maps, the upper showing the sea-level pressure distribution, and the lower showing the instantaneous stream lines at the same time. The stream line map must not be misunder-
and rttindrops from falling vertically, % y the northern
At the outset the following consir P erations were sug-
northerly ap enm, it is possib Y e to approximate the alti- tude of the P reezing temperature by assuming an appro-
alt,itude, w 7l en the wind is bearing a. temperature very
Cha.Tta.- +) o aid in the discussion of the data from the
74 MONTHLY WEATHER REVIEW. FEBRUARY, 19‘20
type of storm has been called by the rather unsatisfactory name of ice s t m n . The ice cover which is formed is
called by the English gZazed rost; this is not satisfactory
the German Glatteis are more descriptive. Our word gluze, adopted by the Weather Bureau in 1916, comes nearest to a satisfactory designation for this condition.‘ The conditions under which such preci itation is pro-
Dr. C. F. Brooks,a and the reader is referred to his article for an explanation of the processes taking place aloft which produce areas of snow, sleet, glaze, and rain, either singly or ih combination. This paper proposes to discuss, as far as available data and legitimate specula- tion will permit, the conditions aloft as well as at the surface, and to show how they interact in a specific
storm.
because it is not related to f rost; the French vergZas and
duced have been discussed in the preceting 7 article hy
PLAN OF THE STUDY.
The winter of 1919-20 has been a particularly unpleas- ant one in the Emtern United States, owing, especially, to the eat precipitation of snow in the northern portions, and o !? snow, sleet, and rain in the middle latitudes, which formed a solid, slow-melting cover. Perhaps the worst eriods of this t pe were t>hose of January 20-25
territor east of the 100th meridian, while the lat8ter was confine i to the Atlantic coast States. The first period has been investigated the more thoroughly owing to its greater mea and duration.
and Fe \ n a r y 3-6,’the s ormer covering practically all the
which so
and glaze formation:
1. Low temperature and high pressure to the northward (between northwestandnortheast.) * * * 2. Steep pressure and temperature gradients to the nurthward (he-
tween northwest and northeast.) * * * 3. Surface temperatures below freezing. * * * 4. Moderately high pressure and high temperature over the East Gulf and South Atlantic States. * * * 5. Northward looping of the isotherms. * * * 6. Gentle to fresh northerly winds, increasing by time sleet begins. * * * 7. Low prmre trough trending southwest to northeast between two
HIGHS. The LOW is uwally moving from the southwest, but sometimes from the northwest.
And, while it is ossible, when such formations are pres-
is rarely ossible to forecast its location more definitely
is, obvious1 , a function of the temperature both at t8he surface angaloft, and of the strength of the north and south components of the underrunning and overriding winds. Even though the critical exmination of a single storm is unlike1 to shed as much li h t on these relation-
ation justifies itself in settin forth clearly throughout
ent, to say that a e eet d l occur or that glaze will occur, it
than to d!) eaignate its-center. But the width of the area
ships as might i e desired, neverthe K ess, such an investi-
%e enod the conditions whic E did prevail together with the Ln d of precipitation and the regions over which it occurred. Meridional sections o f the atmosphere.-Besides the ordi- nary relations of barometric pressure and precipitation, it -__~
1 Abbe. Cleveland. Jr.: American deflllltlon of “&at.” MONTHLY WEATBEB RE- I ~W igie, u: asi-~a: : Brooks Charlea F.: Thenature of sleet and how it Is formed Thls REVIEW pp. 69-72.
*U 8 be of A culture, Weather Bureau: Weather iorecsatinp in {he United W& $6, pp. %7-m.
is necessary to investigate by means of meridional cross- sections the distribution of temperature, cloudine,ss, and wind velocit,y aloft; also to construct maps showing the areas over which precipitation in the form of rain, sleet, or snow, or cnmbinations of these, occurred. I n con-
structing the meridionnl cross-sections two assumptions are nec.essary-first, that the wind of southerly component continues aloft after le,aving the ground, overriding the wind of nort,herly component; and, second, t>hat the tem- perature of the, wind of southerly component ma.y be sur- mised froni the following: (a1 The temperature fall per degree of latitude as t,he southerly wind procee.ds north-
ward can give a clue as to the temperature of the wind afte,r it has risen from the surface; (6) the dynamic cool- ing of the southerly air ns i t rises can be surmised from t,he amount of precipit.ation: (c ) the northern limit of the isotherm of fre.ezing is marked, wit,h due cnnsiderat,ion for the effect of wind velocit,y in keepin the sleet pellets
limit of sleet or rainfall; and (d) the motion, type, and height of the base of the clouds tell much concerning t,he conditions aloft, where kite. observations are’wanting. Width of sleet and gh.zi> belts.-The width of the sleet belt and glaze belt, are, of course, of prime importance in the, forec.ast, and thus the forecast becomes prinisrily one of temperature; i. e., hhe prechting as accurately as possible. the location of the horizontal freezing line. But while this can he n certain guide to the general loca- tion of sleet and glaze, there must be othex conditions peculiar to the winds involved whidi will determine the widt,h of the are,a over which preci itation will occur.
gested: (I) The steepness of the surfnce temperature gradient within the zone of northerly winds; (a) the st,rength of the southward component in the northerly wintls; (3) the strength of the southerly wind aloft; (4)
the difference. between tlie su1.fac.e temperature and the maximum temperature a,loft.; and (5 ) the determination
of tlic height of the maximum temperature aloft. The last point is one that is subject to several considerations.
I€ one considers the surface t)emperature in the southerly wind at some point south of the wind-shift line, or the line where tlie southerly com onent disappears and the
priat’e vertic.al temperature gradient, dependent upon the current conditions. Direct observation by kites is desirable, but where this is impossible, the current condi- tions can be surmised by allowing the vertical temper- ature gradient to depart from the normal shown by kites in accordance with the variation of ot1ie.r condit.ions from the normal, such as wind speed and snow cover. Again, after this southerly current le,aves the ground its temperature must be cl-jmamically lowered. Finall , the strength of the sout,herly component must be consi B ered, for it is ap arent that an increase of wind velocity with
high relative to that at the surface, is certain to carry the warm air farther north over the cold layer than would otherwise be the case. All of these three factors must be included in any speculation regarding the altitude of the freezing tem erature at its northern extre.mity aloft.
various standpoints mentioned above, the accompany- ing series of charts has been devised. The even-num- bered Charts 11, IV, VI, etc., consist of two maps, the upper showing the sea-level pressure distribution, and the lower showing the instantaneous stream lines at the same time. The stream line map must not be misunder-
and rttindrops from falling vertically, % y the northern
At the outset the following consir P erations were sug-
northerly ap enm, it is possib Y e to approximate the alti- tude of the P reezing temperature by assuming an appro-
alt,itude, w 7l en the wind is bearing a. temperature very
Cha.Tta.- +) o aid in the discussion of the data from the
74 MONTHLY WEATHER REVIEW. FEBRUARY, 19‘20
type of storm has been called by the rather unsatisfactory name of ice s t m n . The ice cover which is formed is
called by the English gZazed rost; this is not satisfactory
the German Glatteis are more descriptive. Our word gluze, adopted by the Weather Bureau in 1916, comes nearest to a satisfactory designation for this condition.‘ The conditions under which such preci itation is pro-
Dr. C. F. Brooks,a and the reader is referred to his article for an explanation of the processes taking place aloft which produce areas of snow, sleet, glaze, and rain, either singly or ih combination. This paper proposes to discuss, as far as available data and legitimate specula- tion will permit, the conditions aloft as well as at the surface, and to show how they interact in a specific
storm.
because it is not related to f rost; the French vergZas and
duced have been discussed in the preceting 7 article hy
PLAN OF THE STUDY.
The winter of 1919-20 has been a particularly unpleas- ant one in the Emtern United States, owing, especially, to the eat precipitation of snow in the northern portions, and o !? snow, sleet, and rain in the middle latitudes, which formed a solid, slow-melting cover. Perhaps the worst eriods of this t pe were t>hose of January 20-25
territor east of the 100th meridian, while the lat8ter was confine i to the Atlantic coast States. The first period has been investigated the more thoroughly owing to its greater mea and duration.
and Fe \ n a r y 3-6,’the s ormer covering practically all the
which so
and glaze formation:
1. Low temperature and high pressure to the northward (between northwestandnortheast.) * * * 2. Steep pressure and temperature gradients to the nurthward (he-
tween northwest and northeast.) * * * 3. Surface temperatures below freezing. * * * 4. Moderately high pressure and high temperature over the East Gulf and South Atlantic States. * * * 5. Northward looping of the isotherms. * * * 6. Gentle to fresh northerly winds, increasing by time sleet begins. * * * 7. Low prmre trough trending southwest to northeast between two
HIGHS. The LOW is uwally moving from the southwest, but sometimes from the northwest.
And, while it is ossible, when such formations are pres-
is rarely ossible to forecast its location more definitely
is, obvious1 , a function of the temperature both at t8he surface angaloft, and of the strength of the north and south components of the underrunning and overriding winds. Even though the critical exmination of a single storm is unlike1 to shed as much li h t on these relation-
ation justifies itself in settin forth clearly throughout
ent, to say that a e eet d l occur or that glaze will occur, it
than to d!) eaignate its-center. But the width of the area
ships as might i e desired, neverthe K ess, such an investi-
%e enod the conditions whic E did prevail together with the Ln d of precipitation and the regions over which it occurred. Meridional sections o f the atmosphere.-Besides the ordi- nary relations of barometric pressure and precipitation, it -__~
1 Abbe. Cleveland. Jr.: American deflllltlon of “&at.” MONTHLY WEATBEB RE- I ~W igie, u: asi-~a: : Brooks Charlea F.: Thenature of sleet and how it Is formed Thls REVIEW pp. 69-72.
*U 8 be of A culture, Weather Bureau: Weather iorecsatinp in {he United W& $6, pp. %7-m.
is necessary to investigate by means of meridional cross- sections the distribution of temperature, cloudine,ss, and wind velocit,y aloft; also to construct maps showing the areas over which precipitation in the form of rain, sleet, or snow, or cnmbinations of these, occurred. I n con-
structing the meridionnl cross-sections two assumptions are nec.essary-first, that the wind of southerly component continues aloft after le,aving the ground, overriding the wind of nort,herly component; and, second, t>hat the tem- perature of the, wind of southerly component ma.y be sur- mised froni the following: (a1 The temperature fall per degree of latitude as t,he southerly wind procee.ds north-
ward can give a clue as to the temperature of the wind afte,r it has risen from the surface; (6) the dynamic cool- ing of the southerly air ns i t rises can be surmised from t,he amount of precipit.ation: (c ) the northern limit of the isotherm of fre.ezing is marked, wit,h due cnnsiderat,ion for the effect of wind velocit,y in keepin the sleet pellets
limit of sleet or rainfall; and (d) the motion, type, and height of the base of the clouds tell much concerning t,he conditions aloft, where kite. observations are’wanting. Width of sleet and gh.zi> belts.-The width of the sleet belt and glaze belt, are, of course, of prime importance in the, forec.ast, and thus the forecast becomes prinisrily one of temperature; i. e., hhe prechting as accurately as possible. the location of the horizontal freezing line. But while this can he n certain guide to the general loca- tion of sleet and glaze, there must be othex conditions peculiar to the winds involved whidi will determine the widt,h of the are,a over which preci itation will occur.
gested: (I) The steepness of the surfnce temperature gradient within the zone of northerly winds; (a) the st,rength of the southward component in the northerly wintls; (3) the strength of the southerly wind aloft; (4)
the difference. between tlie su1.fac.e temperature and the maximum temperature a,loft.; and (5 ) the determination
of tlic height of the maximum temperature aloft. The last point is one that is subject to several considerations.
I€ one considers the surface t)emperature in the southerly wind at some point south of the wind-shift line, or the line where tlie southerly com onent disappears and the
priat’e vertic.al temperature gradient, dependent upon the current conditions. Direct observation by kites is desirable, but where this is impossible, the current condi- tions can be surmised by allowing the vertical temper- ature gradient to depart from the normal shown by kites in accordance with the variation of ot1ie.r condit.ions from the normal, such as wind speed and snow cover. Again, after this southerly current le,aves the ground its temperature must be cl-jmamically lowered. Finall , the strength of the sout,herly component must be consi B ered, for it is ap arent that an increase of wind velocity with
high relative to that at the surface, is certain to carry the warm air farther north over the cold layer than would otherwise be the case. All of these three factors must be included in any speculation regarding the altitude of the freezing tem erature at its northern extre.mity aloft.
various standpoints mentioned above, the accompany- ing series of charts has been devised. The even-num- bered Charts 11, IV, VI, etc., consist of two maps, the upper showing the sea-level pressure distribution, and the lower showing the instantaneous stream lines at the same time. The stream line map must not be misunder-
and rttindrops from falling vertically, % y the northern
At the outset the following consir P erations were sug-
northerly ap enm, it is possib Y e to approximate the alti- tude of the P reezing temperature by assuming an appro-
alt,itude, w 7l en the wind is bearing a. temperature very
Cha.Tta.- +) o aid in the discussion of the data from the
74 MONTHLY WEATHER REVIEW. FEBRUARY, 19‘20
type of storm has been called by the rather unsatisfactory name of ice s t m n . The ice cover which is formed is
called by the English gZazed rost; this is not satisfactory
the German Glatteis are more descriptive. Our word gluze, adopted by the Weather Bureau in 1916, comes nearest to a satisfactory designation for this condition.‘ The conditions under which such preci itation is pro-
Dr. C. F. Brooks,a and the reader is referred to his article for an explanation of the processes taking place aloft which produce areas of snow, sleet, glaze, and rain, either singly or ih combination. This paper proposes to discuss, as far as available data and legitimate specula- tion will permit, the conditions aloft as well as at the surface, and to show how they interact in a specific
storm.
because it is not related to f rost; the French vergZas and
duced have been discussed in the preceting 7 article hy
PLAN OF THE STUDY.
The winter of 1919-20 has been a particularly unpleas- ant one in the Emtern United States, owing, especially, to the eat precipitation of snow in the northern portions, and o !? snow, sleet, and rain in the middle latitudes, which formed a solid, slow-melting cover. Perhaps the worst eriods of this t pe were t>hose of January 20-25
territor east of the 100th meridian, while the lat8ter was confine i to the Atlantic coast States. The first period has been investigated the more thoroughly owing to its greater mea and duration.
and Fe \ n a r y 3-6,’the s ormer covering practically all the
which so
and glaze formation:
1. Low temperature and high pressure to the northward (between northwestandnortheast.) * * * 2. Steep pressure and temperature gradients to the nurthward (he-
tween northwest and northeast.) * * * 3. Surface temperatures below freezing. * * * 4. Moderately high pressure and high temperature over the East Gulf and South Atlantic States. * * * 5. Northward looping of the isotherms. * * * 6. Gentle to fresh northerly winds, increasing by time sleet begins. * * * 7. Low prmre trough trending southwest to northeast between two
HIGHS. The LOW is uwally moving from the southwest, but sometimes from the northwest.
And, while it is ossible, when such formations are pres-
is rarely ossible to forecast its location more definitely
is, obvious1 , a function of the temperature both at t8he surface angaloft, and of the strength of the north and south components of the underrunning and overriding winds. Even though the critical exmination of a single storm is unlike1 to shed as much li h t on these relation-
ation justifies itself in settin forth clearly throughout
ent, to say that a e eet d l occur or that glaze will occur, it
than to d!) eaignate its-center. But the width of the area
ships as might i e desired, neverthe K ess, such an investi-
%e enod the conditions whic E did prevail together with the Ln d of precipitation and the regions over which it occurred. Meridional sections o f the atmosphere.-Besides the ordi- nary relations of barometric pressure and precipitation, it -__~
1 Abbe. Cleveland. Jr.: American deflllltlon of “&at.” MONTHLY WEATBEB RE- I ~W igie, u: asi-~a: : Brooks Charlea F.: Thenature of sleet and how it Is formed Thls REVIEW pp. 69-72.
*U 8 be of A culture, Weather Bureau: Weather iorecsatinp in {he United W& $6, pp. %7-m.
is necessary to investigate by means of meridional cross- sections the distribution of temperature, cloudine,ss, and wind velocit,y aloft; also to construct maps showing the areas over which precipitation in the form of rain, sleet, or snow, or cnmbinations of these, occurred. I n con-
structing the meridionnl cross-sections two assumptions are nec.essary-first, that the wind of southerly component continues aloft after le,aving the ground, overriding the wind of nort,herly component; and, second, t>hat the tem- perature of the, wind of southerly component ma.y be sur- mised froni the following: (a1 The temperature fall per degree of latitude as t,he southerly wind procee.ds north-
ward can give a clue as to the temperature of the wind afte,r it has risen from the surface; (6) the dynamic cool- ing of the southerly air ns i t rises can be surmised from t,he amount of precipit.ation: (c ) the northern limit of the isotherm of fre.ezing is marked, wit,h due cnnsiderat,ion for the effect of wind velocit,y in keepin the sleet pellets
limit of sleet or rainfall; and (d) the motion, type, and height of the base of the clouds tell much concerning t,he conditions aloft, where kite. observations are’wanting. Width of sleet and gh.zi> belts.-The width of the sleet belt and glaze belt, are, of course, of prime importance in the, forec.ast, and thus the forecast becomes prinisrily one of temperature; i. e., hhe prechting as accurately as possible. the location of the horizontal freezing line. But while this can he n certain guide to the general loca- tion of sleet and glaze, there must be othex conditions peculiar to the winds involved whidi will determine the widt,h of the are,a over which preci itation will occur.
gested: (I) The steepness of the surfnce temperature gradient within the zone of northerly winds; (a) the st,rength of the southward component in the northerly wintls; (3) the strength of the southerly wind aloft; (4)
the difference. between tlie su1.fac.e temperature and the maximum temperature a,loft.; and (5 ) the determination
of tlic height of the maximum temperature aloft. The last point is one that is subject to several considerations.
I€ one considers the surface t)emperature in the southerly wind at some point south of the wind-shift line, or the line where tlie southerly com onent disappears and the
priat’e vertic.al temperature gradient, dependent upon the current conditions. Direct observation by kites is desirable, but where this is impossible, the current condi- tions can be surmised by allowing the vertical temper- ature gradient to depart from the normal shown by kites in accordance with the variation of ot1ie.r condit.ions from the normal, such as wind speed and snow cover. Again, after this southerly current le,aves the ground its temperature must be cl-jmamically lowered. Finall , the strength of the sout,herly component must be consi B ered, for it is ap arent that an increase of wind velocity with
high relative to that at the surface, is certain to carry the warm air farther north over the cold layer than would otherwise be the case. All of these three factors must be included in any speculation regarding the altitude of the freezing tem erature at its northern extre.mity aloft.
various standpoints mentioned above, the accompany- ing series of charts has been devised. The even-num- bered Charts 11, IV, VI, etc., consist of two maps, the upper showing the sea-level pressure distribution, and the lower showing the instantaneous stream lines at the same time. The stream line map must not be misunder-
and rttindrops from falling vertically, % y the northern
At the outset the following consir P erations were sug-
northerly ap enm, it is possib Y e to approximate the alti- tude of the P reezing temperature by assuming an appro-
alt,itude, w 7l en the wind is bearing a. temperature very
Cha.Tta.- +) o aid in the discussion of the data from the
I
-. ' ---ill ,..,,. , .... r:---. . . ,
I
..-
I
Y -
L-
- .-
--
- .-
--
-
..
1v
I ?Irm. I Ill
INTERMEDIATE CLOUDS
WIND BOUNDARY --__
/
:z } D
WIND EOUNDAR
Surf.ce
-
..
1v
I ?Irm. I Ill
INTERMEDIATE CLOUDS
WIND BOUNDARY --__
/
:z } D
WIND EOUNDAR
Surf.ce
I
I
I
I-
I
February, 1920. M.W
:-
42km. I
.. . ... . P: ,. , .. .>. . ., . . ..* ..
_j
.. . .
I
FEBRUARY, 1920. MONTHL1- WE+THER REVIEW. 75
stood: it! does not show the actual patlis of particles-of air,. but, on the contrary, merely the tlirect.ions with which the winds were hlowino nt t7t.e time qf oI)seiwti.on..
This method has been devisedxy Bjerknes m t l lins becn used by him in liis stutlies in forccast,ing in Norway and other arts of the world. Our maps also show, wlicrevcr
onent after it litis left the pourid n t tile wincl-shift lint.. Fn these maps the suiqz , c(.! winds itre sliown in red and the robublc cou1*3c of the soiit,licrly wintl uloft. by tlot.tctl Hnes . Opposite tliesc maps, for :t gjveri time, is the mrip sliow-
ing the distribution of precipit:i.tion of vttriou3 k i d s during the preceiling 12 hours. This shows, in overlap- ping areas, the clinrncter of t?ie prwipi tiition. By plac-
ing the isotherms on this mnp we able to sliow tlic
region which probably includes the iirc..~s ovcr which glaze formed. namely, all that arm north of the fretmiip
h e (33O F. isotherm) ovcr which r i n fell during tlic
In addition to the isotlirrni of
Freezing, which, owing to it,s importnnce is more promi- nent than the otlimv, the isotherms of ?so, Wo, 10° (F.1 etc., are marked north of the frerzing liiie, riiitl -H I o , SOo
(F.), etc., are marked south of the freezing line. On the lowcr part. of the precipitnbion ~n i l tein rniture map, we have )repwed sections of t l i c a lowcr 3 ki P omet,iim
of the atmosp f l ere along selccteil meritli:ins. The meri- dim selected is note!{ by 11 henvy h1xc.k line on t.lw mnp, and its c.orresponding section below is found by m:ms of
the index letter. These sections arc! the result oi putting together those fragmentary tlrita wiiicli, owing to tlw
regrettable sciircit,y of ucrological stnt ions in t.lw L'nitail States, consist. of isoliitecl t.empernturc* mensuremeiit.s a t
various levels in the free air, and of wind clirectioiis :inti
speeds, am lifietl with speculutions which nre based upon previous o E servations. Tiius, while tliesrb wrtions (:nu
not lay claim to being actual representittions of the con- ditions aloft, they may be regarilctl with consiclariible confidence. These maps have bocn prepared for evt?ry 12 . liorirs
throu hout tlie period under invcst.ipntion, hegmning with 0 anuary 21, S n. in. The lnst two maips (Chart XVIII) show the ressure and t,emperaturc rontlitioris
eriod with tliab of t,lie 27th to 31wt clisrusseil in thc
Following article in this REVIEW.
possib P e, the probable path of the wiiitl o f sout.lierly coin-
receding E houm
on January 35 mi f 26, to coniirct the wc1:itlirr of this
GENERAL SURVEY OF THE PERIOD.
Presmre.-The period opens (C11art 1, priwuro :inti
wincl-shift are noted on s:inie m n p with ot.lier rlatii for January 30, 8 p. m.) with the territory uncler consi!Icrr+ tioii flanked by high pressure. in the northwest., nnrtli- east, and southeast, with moilcrate clc ressions in t.hc Lalie region ant1 in t\iv Southwest. !hie sul>scclucbnt.
:i genernl brad the northern p:trt. of was n general tlecreaw
toward the Gulf which was manifcstrtl in the Gulf States by a fairly persistent protubermce of lower ressure which estended from the; Gulf to western North 8aroliau snd Virginia (Chart IVit). The genernl urea of high rcssurc off the soutlieast const wt1.s also quitc pe!sist,ent. bn the morning of January 23 II, small depression was centered over northeastern ibk:instis (Chart Xa) mi I by evening of the same t h y wis t~lliptictil in form with
its major irxis roughly parallel to thc Appalachian Moun- tains. With the strengthening of the high pressure in the north and the appenrance of a center in northern Minnesota on .January 24 (Chart SVIa), the isobars 8,s-
sumed I L southwest-northeast trend with the protuber- ance of low pressure, previously mentioned, lying along the south At,lnntic coast.. 'I'his was followed, coinci- dently with the general disuppenrance of sleet conrli- tions, by t i n ~nsterly motion of the HIQH to the Lake region, :I.I~I-I the southeastern LOW to Floritln (Chart SVIII), on the morning of January 25.
What is the significance of this pressure tlistributiou, 1rit.h res wc.t to tlie coiiilitions favorable for sleet and glaze.? fire mentioned above that, oiie of t.he major
premisrs of a sleet. forecast is the prestmce ill the sou the& of a HIGH, the temperatures wit.liiii H hich are not so low
as usual, just off t.he coast of southeastern Iinitecl States. Suc.h ii condition persisted uiitil the last few hours of
t,he ieriocl. Morcorrr, t.he belt of high pressure, evi-
au ,ply of cold air, sweeping down from tlie deep snow-
griiernl area of low pressure along the soutlierii coast.. llic returii hir of w\'c~rni moist air, riding over the wedge- like c~ncroac.Iiing north wiiid, was cooled and forced to precipit.ate raiti, sleet,, or siiow over niost of tho eastern I'iiited St.at.cis ~;Iirougliout. t l ~e period. W/d.---.?Jow, let us look to the charts of instantaneous stream 1iiii.s of surface wiiid. Chart. I shows a strongly niarkcd witid-shift line tlieavy dotted red line) cxtendmg from Tcsiis iiorthwwterly toward the center of low
j)ressurt~ ovcr IJcike Eric.. South of t.his linc sll tlie wiiids, with wrtaiii usccptioiis iu the mountain district where topography prevents tlie surface air from flowing readily wit,h t.he geiiersl wind riot far aloft, liave u south com- porient. (Iildicutecl on Chart I by heavy red arrows.) All winds iiortli of this line liave a 1iort.h component. Chart IIb shows the sanie wind-shift line extending from emt Texas to New England with a slight. break in North Carolina, owing to the slight depression cmtered in t.hat re rioii. Throughout the period, it is possible
in Chart. VIb, for instance, but reco~ering its identity a few hours latcr LS i t i Chart. S b . As the end of the period approached , the sweep of the northerly wiiicl became reater until finally it covered t.he Gulf and South %tlaiit.ic coasts. (C1inrt.s SIVb, SVIh, SVIIIa.) Tmyeratwr n .~t ? 2"'eciPihti.ori.--Thc picture of the current w-cather is completed, as far as our interests are
coiicerned, by t,he odd-numbered charts, which show the tcniperature at. tlie surface and the areas over which various k i d s of precipitation have occurred in the pre- ceding 12 hours. Briefly, attention is called to the general as1)ect.s .of these chart,s. The period begins with geiicd prrcipit.at,ioii west of tho C6th nieridian. The
area of glaze corers Illinois and the norbhern portions
of Indiana sud Ohio, and again in the east, a corisiderable portion of the Middle Atlantic States. Sleet also fell over
a large area from Lake Michigan eastward to the Atlantic aiid as far south as Philadelphia. The following 12-hour
shoivs the major ortion of the sleet area to lie in
uith another, srualler area in Pennsylvania and New York. .Attention, however, should he called to the fact that the maps represcntiiig recipit.ation during the night hours
is believed that this is attributable to the lack ofdose and careful observation during the night hours. Eren during
dent l y cetitcred far north 111 Canada, gave an abundant
tie I CIS of the north far into the United States toward the
r 7
to see this \ iiic, sometimes rather indefinitely located, as
Fd issouri and sout.lieni 8 liiiois and Indiana (Chart III),
seem t.0 sllow a tnnt f ency toward spottedness of sleet. I t
76 MONTHLY WEATHER REVIEW. FEBRUARY, 1920
is an easy matter to fail to note sleet when small quantity with rain or snow, and the only increase the difficulty. This, combined with the less vigilant night observations, might account for the spottedness of sleet areas. The general trend of the isotherm of freezing throughout the period is from southwest to northeast and the rain areas conform in general to the same direction, the rain belt estending roughly equal distances on either side of the isotherm. As the period draws to a close the areas of sleet and gla!t! growii smaller and, while large areas of snow and rai!i are still prevalent, the sleet and glaze producing condi- tions are disappearing. Another outstanding feature of thc temperature dis- tribution is the distinct northward hook in the mountain region, the line of freezing often proceeding far north of its general trend because of the inversion of Eempera- ture which resuIts from the overrunnin warm wind from the south. The lower stations, there P ore, are in the northerly wind, but those of greater eleration project up into the warmer region and swing the isotherms far froni their normal course, i. e., the course they would probah1.y pursue if the mountain region mere absent. Tlila mversion is nicely shown on Chart VII in which Elkins,
W. Va., shows a temperature several degrees higher than the territory in its immediate neighborhood. Toward the end of the period, with deep northerly winds, the opposite is the case (Chart SV), for Elkins has a tem- perature considerably lower than that of the surrounding country.
THE MERIDIONAL SECTIONS.
ExunvpZe.-In order that the significance of the small meridional cross-section maps may be thoroughly uncler- stood, the method employed in drawin5 them will be followed through for one case, which is fairly complete in details. Not all of these sections are of equal weight to the lack of upper-air data in mang cases, but OlvinE t e reasoning employed in their construction IS sonie- thing as follows: Let us take Sectioii A on Chart 111, which represents the conditions along the 83d meridian at 8 a. ni., 75th meridian time, January 31. The general nature of the topography is indicated in solid black. The surface temperatures, that is, the points where the \yarious isotherms of the map above intersect the black h i e All, are indicated along the "surface" line. From Chart Ilh, on the opposite page, we find that the wind-shift line crossed this meridian in northern Georgia, which is to say that all winds south of the line had a southerly com- ponent and all winds north of the l i e had a northerly component. Moreover, that same morning at Leesburg, Ga., a kite ht was made. The conditions there were
hi5h as 3 kilometers. The surface temperature was 57 F.; and at 1, 2, and 3 kilometers were 52", 5l0, and 41°, respectively. The record showed a layer of stratus cloud havin its base at about 1,200 meters, and its top
about 1,500 meters. Within the cloud, the temperature fell to 50°, which is approsimately according to the retarded adiabatic rate. Immediately at the top there was a sudden rise to the 54O observed at 1,500 meters. .Above that level the temperature fell at the greater rate of about 1" F. per 100 nietem5
fog at the su 3 ace and winds of southerly component as
as indicate d by the hygrometer and thermograph, at
* Unfortunately, it was necsssarv owing to the manner in which data are presented by Weether Bureau stations, to emp oy !wemperatures and pressures in FBhrenheit d e w and inches, respxtipely and altitude in meters. But smca the mixing of the s stems d nnits in this caw, do& not result in ccmfusion the author has refrained from de con- vermbf brge numbers of observations to met& un~ts.
At Lansing, Mich., where n pilot balloon nin was obtained, winds of norther1 component up to 2 kilo- metem showed clearly that t t e wind-shift line which was on the ground in northern Georgia must lie somewhere aloft in t8he intermediate distance. Royal Center, Ind., which is in the vicinity of the meridian in question, showed a northerly component at 1 kilometer and clouds at about 1,500 meters. What clue can that 've to the location of the wind-shift line'? It is likely a at at tlie boundary between the winds of north and south compo- nent, there will be clouds formed by misture, that is, the mixinw of the cold air from the north with the warm, moist air %om the south, will produce a re on of inter-
moisture content, with the result that cloud will form. Thus, if we find cloud at 1,500 meters abo.oe Ro a1 Ceu-
com onent; hence, we will be justified in joining the
mediate temperature, helow the dew point P or air of that
ter, we are robably not f a r wron in saying t < at that is the houn a ary between the win f s of north and south
win B shift a t trhe surfncc with this point in the free air and
Now, let us consider the vertical distribution of tem- erature, which mny be well surmised, knowing the sur-
!ace horizontal distribu'tion and the upper-air conditions with respect to height and thickness of clouds, and wind direction. B inning with the 32" isotherm we find that i t lies well wi%in the north wind, thus the likelihood of a temperature inversion is small and if it did csist, it would not be very strong; hence, we can assume the tem- adient up to the cloud base is the normal for perature winter. % ut as soon as the south wind is encountered
tation, shown on the map from the vicinity of the wind-shift line a t the surface in Georgia to central Ohio. About half the rain mea, along the 83d meridian, lay north of the surface freezing line. But we know that if rain falls u on earth below the freez- ing temperature, there must !e a higher temperature aloft in which rain ma form or in which the snow will
be above freezing. So we C ~T Y our freezin line from the oint where it enters the cloud close1 on the wind
strictly, the northern limit of rainfall of the preceding 13
hours.
This brings to li h t the fact that the sections are not
aloft is really its northern extreme during the preceding
melt as it falls throug z . Such a layer must, obviously,
s nchronous throug 5 out, but that the surfme location of t K c 32" isotherm is current, whereas its northern limit
!ounda,ry to the northern limit of the g yf me be K t, or, more
FEBRUARY, 1920. MONTHLY WEATHER REVIEW. 77
12 hours. Sleet is formed by partially (or totally) melted snow falling into a cold layer and refreezing, or by the freezing of raindrops in the air; but since no sleet fell on on this meridian we should be led to think that the thick- ness of the layer whose temperature was above freezing e, since no snow, which may have fallen from
melted. Curving the freezing line back at this point, we start it in a southerly direction again a t the retarded adiabatic rate. Taking the 40" isotherm, in a similar manner, we raise it a t a normal rate to the cloud level, then sharply northward, as i t encounters the southerly wind, but not so far as the 32" line. As it rises through the cloud it follows the retarded adiabat.ic rate until it reaches the cloud top, where there is a sharp inversion (mentioned before in connection with the observed 51" at Leesburg), a.nd then it proceeds a t a more rapid rate, say
1" F. per 100 meters. We have then, the picture of the processes at work in producing laze formation, the underrunnin north wind,
few kitme and pilot balloon observations, in piecing to-
ether a cross section of tho phenoniena of the upper air. %ot all the sections will be so complete. Some ttre deyoid utterly of assist.ance from the iipper-air observations, because there were no stations new eiiough whose obser- vations could be considered IS applicable to the meridian in question. This detailed esplanation of tlie method of constructing a cross-section shows what ccinsidern t.ions were given to each of the others, to a greater or less de ee.
F-4 in.terestin.g fea.tsire.s.-.As one follows through tlie sections there are a number of points whic.h are brou h t to notice in a very striEng manner. nmong t,lle frst of these is Section C, Chart 111, which shows the effect of gentle souther1 winds, ascending over a wedge of northerly wind w I ich is apparently descendin . This fact is ascertained from the kite observations a t E lendale,
N. Dak., in which we find a wind of northerly component at 2 and 3 kilometers, but of southerly component below
1 kilometer. It is obvious from the stream-line map (Chart IIb) that the southerly surface wind at Ellendide can not be related to the southerly wind of the more southerly latitudes. Hence, we must think of tlie forma- tion as a series of layers of different direction sliding over one another. The wind-shift line or wind boundary probably does not extend far north as we have assumed others to do, but is prevented by the superior strength of tlie northerly wind. A second wind boundary, unrelated to our investigation, would be that between the southerly and northerly winds above Ellendale. A similar phenomenon appears in Sect,inn C, Chart V, in which the northerly component. is introdrircd by u.
northeast wind which, at ita masimum, is perhaps not orer a kilometer and a half in depth. This forces itself like a wedge from tlie side under the southerly wind, lifting it in a broad hump, but dowing the sciiitherly mind to meet the surface again in the north. This is
shown by kite flights at Ellendale, which seem to indicate that an adiabatic temperature gradient preTails above that. region. In tlie south, however, where the isotherm of freezing is t,o be fonnd, the sky is pmtlv cloud^-, rnltrking the nnrthern edge of t,he cloud levej, but the clouds are moving from t.he south, while just below that level, a pilot balloon obserT-at.ion a t Broken Arrow, Okla., discloses a wind of northerly conipnnent. Aiio ther feature of interest is t.hat the precipitation of the pre- ceding 12 hours occurred all on the windward slope of
the wind hump, just as it would on a topographic hump,
higher was '7 evels, was able to get through only partially
the overrickg south wind, the immense va P ue of wen a
P
where a passing moist southerly wind would hme been cooled by expansion rising 011 the southern slope, making
the northern slope, tending to evaporate the cloii 8 s. down precipitat.ion, but warmed by compression flowin
A warning should be made here. The section w e are discwssing shows the northern cloud limit considerably south of the northern limit of ra.in as shown in the sym- bols beneath, and it is desired to menticin again that the preci litation refers to t,he preceding 12 hours, while the rlidiness, tcniperature, etc., are current. Confusion in t,rying to reconcile recipitntion wit.h a clear sky will be avoided if this is g oriie in mind. Mect.ion A, Chart V, sliows the northern limit of rain
and sleet roiiiciding with the masimum altitude of the land. This is shown also in Section A, Chart VI; See- tion A, Chart, IS and Section A, Chat, SI. There are probably two reasons for this phenomenon, operative in combination. First, IS the soiitherlp wind ap roaches t,he mountainous regions i t is deflected upwarl by the land itself, ct1i;sing precipitation from clonds formed by €o r d ascent: second, it is possible t,hnt rain is frilling h n i the climd all dong, but that t.he swface t,empera- tiircs in the more soi:therly rcgion tire sufficientJy high to
ei--a.l)orntz the falling ruin More it reaclm the surl'ace.
h i t as greater devatioiis a.re nttained in the topography, and RS the swfncc tempcrat.ure is lower in more northerly lntihdes, rtiin r e a c h t.lie surface, chicfly hecaiise the rnrt,h goes i!p to meet the rain.
Oiic will notice on Chart SIII, Section B, that. tliere
was a narrow region where slcct was reported in com- hination wit.11 snow, and snut,h of the region of sleet and snow t.lirre is an area where snow alone fell. It seems nutiiral in the ideal sec.tion of the at.niosphere that the scqi!encle o f prwi 1it.at.ion from t,he soi! t,li dong a given meridian should b e rain, rain a.nd sleet., sleet,, sleet and snow, snow; and, in general, such a sequence is found. But here we hare an area of sleet falling in t.he midst of
n snowstorm. The sky is overcast wit.h low clouds €rom which snow is falling. Surely there m i be no ternpya-
t.ure of freezing or above below thnse cl0urls and 111 a
northerly wind. We miist draw npon ow imagination
to find the soiirce of the sleet. We know that it may start as snow and fall through a layer iif air whose tem- pcrat~ire is above freezing. Therefore we may suppose that. tliere is a. cloud laver precipitating snow at snme elevation above the tongue! of southerly air, and that either this cloud area is sinal1 because the area of slmt on the ground is small, or that only a. smnll portion of it is vnder-run by warm air. Hence, we may mark a cloud
Iayer at roughly 2 kilometem But. tho presence of this sleet, area is made more interesting bv a.n observation made earlier in the dav a t Lersburg, Gn.
L; on Janiiary 23 (chart, XI, See. B)? after a morning
fog gad cleared, a layer of [alto]stratus clouds a t about
2 kilometxm was seen moving rapidly northward. Low c*louds prevented us from knowng how far north this layer estended. The wind was of considerable velocity from the soiit,h, and if we may assume that tho layer estended for a considerable distance north of the place of observation, it is quite rwnr.einible that the clouds whkh produced t,he sleet in Section B, Chart XIII, are
t.he same as those headed rapidly northward from Lees- biwg. This is not a far-fetched conclusion, but is an in terestin piece of aerological detective work.
Cbart VII. Here there is no wind-shift line; the surface wind throughout the length of the meridian is northerly. There are no kite sttitions within the immediate vicinity of this meridian, nevertheless, those at Drexel, Nebr.,
A fourt a point worthy of attent,ion is seen in Section B,
78 ’ MONTHLY WEATHER REVIEW. FEBRUARY, 1920
show a wind of southerly component and of t.emperatnrrs c.onsistent wit.11 those at the surface and with the area
over which sleet fell. It is therefore ])robable that over t.his whole region there wils only a sht~llowv nort,h wind
niirl n south wind aborc it Id consiclcmble ilepth. Finally, mention should be macle of the apparcw t dis- continuity bcbween t.hc arms of rainfall and ?rnowfall. ‘I C s r q t for t.hose perids when prccipit.ltt.ic ~I I WBS plneral tlvoughout the eastern half (i f the United Stnt,es, as in Chart 1111, therr nppenrs a. well-dqfinccl clear spawe 011
the niaps in w1iic.h no recipitat.ic.iii occmrecl lyinp
how the snc.)wlnll WLLS prc)bnhly 2tth-i h1itdde to IL ~~.i w
quitc dist.inct, from that which \vas producing t,lie rain and sleet farther svi.itli. For example, take Charts VZI.
IS, and SI. in which there is shown tho advanw from t.he northwest (first ajqmiring in North Dakota 011 (.iiir
c.liarts) c-jf an area snowfaLl. first, only B small area being risible, then widening m c l advancing, and, finally, jniniiig itself to t.he general preripitation urea. Thc
cause of this is probrtbly tis ~ollo~vs: X q h c e st thti pressure map 1% shows that. a high-pressure arm WM
a >pearing in this rcsion, and thut suow was falling where
t,ke pressure gradient wm steepest, the previpitdion apparently being the result of forced ascent of tlie air
as it. flowed into this belt Froin the rwr :it) L highrr velocity than it flowed out in frciiit.
between the miu area mi( r tlie snow wen. This show;.;
THE S ‘~0 l t M OF FEBRII’AHP 3-6, 1W30.
While the precipitation of sleet mcl the formation of
glaze during the period of Febivary 3 to G was more local in charncter and somewhat shorter in duration than thc one we have just disvussnd, it may he safely saic! that it stands out in the minds of those who experienced it as one of the most severe storms in years. t occurred chiefly
east of the Appalachian Mountains, hence, because of the relative1 small area covered we haqe not c.onsidered it in the 2 etailed manner of the prevlous storm, and no oharts have been drawn for it. The conditions in New England have had a parallel only in three or four in- stances. Trains were stalled for da s, roofs cdlupsed from the weight of snow and ice, a d i n New Yorli City the removal of the snow and ice afforded the most per- plesing problem of its kind in the city’s history.” The great mass of cold air over the snow-covered north, with its attending area of estremely high pressure, which lay over the North Atlantic States, afforded a strong bar- rier to the cyclone which, having appeared in the south- eastern States, could move but slowly northward, and, therefore, increased markedly in intensity. The circula-
tion about this strong csyclone, in co1r;lbination with the
HIGH in the north, served to send winds of gale force, sweeping from the northeast down the coast for several days. Above this wedge of biting wind there flowed a layer of moisture-laden. air from the balmy espanses of the Gulf Stream, and in come uence there was heavy
8ew England, New York, and Pennsylvania, with snow, sleet, and glaze-forming rain along the coasts, while farther south therc was an unusual quantity and dura- tion of sleet and rain. On February 3 the cyclone was centered near Savannah. The wind-shif t line lay across northern Florida, thus allowing estremely warm air from Florida and the Gulf to ride UJ-, over the cold air fed by the strong wall of high pressure in the Lake Region and New England.
recipitation of snow (several Y eet) in the interior of
.
6 See note on “Demmalirotlon of trofRc In New York City by snow and sleet,” this REVIEW, p. 80.
On February 4, the center of the cyclone had moved out to sea from Savannah, leaving the wind-shift line sliphtly farther north than on the previous day, but still we31 marked in southern Georgia and northern Florida. The c clone apparently changed its course on February 5 and z egan t o move roughly parallel to the coast in a northeasterly direcation. The wind-shif t line was no longer in evidence, and the horizontal surfacc temperature gradient was very slight, although the temperatures were rather low. Precipitation during these days was general along the coast, with lieav snows in the north, which
and glaze were predominant, the sleet falling in such quantity at Washinpton, for instance, as to afford a cover several inches deep much like dry sand. Rain, of course, fell in tlie South Atlantic States. The conditions producing this storm were obviously of the same general nature as those of tho previous one. escept more intense and more localized. So much warm air blew inland during this storm, that after the storm passed off, tho northwest wind WIW relatively mild in temperature.
were mixed wit,h rain and s r eet. Farther south the sleet
PRECEPTS FOR FOREObgTING SLEET AND OLALE.
It was stated in the kginning that tlie motive for pre- paring this paper was t h desire to find some incans to
improve the forecasting of sleet and glaze: at the same time it was observed that no one can well anticipate the clisrovery of n panacea for all the difficulties of this type of forecasting from the study of n single storm. There
are, howrver, wrtnin features of this storni which appear as c!haracteiiutic, i. e., they seem to persist, as long as the sleet and glaze persisted, and, therefore.. are probablv rellated. One of these features is the ptwiillellism between the 3Zo isotherm, the wind-shift line, and the northern limit of rain and sleet precipitaDion. While the estrerne value of forecasting the location I2
or 24 hours later of the 82’ isotherm has been pointed out in a prcrious reference to the statement of Dr. Franlr- enfield as to the conditions which procluw sleat and “ice storms,” it appears that it is also of estreme importance to forecast, and pay c uite as great heed, to the location of the wind-shift line. kor the two are .inseparably related in the production of this troublesome type of recipitation.
which glaze mill form, and the area over which sleot will fall. Heretofore, the centers of these regions have ?wen forecast with fair accuracy, but often the forecaqting of the limits has been difficult, if not im ossible. It he- hooves us, then, before this paper can xe completed, to investigate the question of whether or not there is any relation between the distance from the 32’ isotherm to the wind-shift line and the values in question, namelv, the width of the giaze belt, the width of the sleet hdt, and the distance of the center of the sleet belt north of the 32O isotherm. Consequently, these distances have bcen measured throughout t4he period along various meridians, and ratios have been formed which may serve as factors in predicting the desired values. The measures were made on 34 meridians during the storm, with no attempt to select specific conditions. In spite of the fact that, the espccted large divergences from the uveragc do occur, it is surprising, nerertheless, that t-here is as
good apecment, as IS found. The rnlues which have been determined are presented with the full recognition that their determination htls h e n purely empirical, and with the warning that they should be considered in the same cautious manner in which thoy are given, since they have been determined from this single storm, and
The forecaqter’s desire is to forecast, t % e area over
F'EBRIJAEY, 1920. MONTHLY WEATHER REVIEW. 79
may be found, when compared with other storins, to show a wide divergence. We do not ignore the upon which swh re- lations, as we seem to empiricd measure-
ment, may rest. The t,he 32' isotherm and the wind-shift line is taken as the basis, and the fore- casting of the location of each of these is dependent upon the pressure distribution. But the 32' isotherm is usually found in this type of storm to lie north of the wind-shift line; that is, within tshe area over which wind of northerly com Ionent is flowing, and warming as it reaches the south- erly 1 atitudes. The wind-shift line marks the place where the southerly wind leaves the aurftice and moves northward aloft. Kites and pilot balloons will show how far nvrtli this wind has progressed, providing there are 811 ough such ol>servations. The amount of precipitation will depend upon the moisture content of the southerly air. The ultimate estent of the southerly wind aloft will dt?peiicl upon its speed, and, similarly, the ultimate extent of the northerly wind a t the surface will depend upon it,s
speed and the nature of the surface over which it is How- ing. These are general considerations which the Iwe- caster is accustomed to make, and which he can makc with inc.reasingly great certainty as he is armed wit,li more and more data and obserrations from kites, pilot balloons, and clouds. Bearing these things in mind, the empirical vulues re-
ferred to above are now presented, with a rough inciuure
of the certainty or accurac.y of the occurrelice (I f the Dhenomenon :
Width of t.he glaze belt. (" Distance between 3 3 O isot.herm and
= 0.7 Width of the sleet belt.
Distance between 32' isotherm and wind-shift l i G (2)
Distance between 32' isotherm and center of sleet.
Distance between 33' isotherm and wind-shift line. = 0.8
The natural question is, With what accuracy cln 1i-e
know these values? It is obvious t\hhnt in n proposition of this nature, with only 34 determinations, it is unwise
to attempt to applv any st,atisticnl fni.mulae for the es-
pression of the prohable accuracy. But of the measure- ments made from the maps, we may make the following st a temen t :
11 agreed within 20 per cent of the mean.
22 agreed within 40 per cent of the mean.
29 agreed within 60 per cent of the mean.
8 agreed within 20 per cent of the mean. 15 agreed within 40 per cent of t,he mean.
25 agreed within 60 per cent of the mean.
7 agreed within 20 per cent of the mean.
12 agreed within 40 per cent of the mean. 18 agreed within 60 per cant of the mean.
One can judge that the accuracy is only sufficient to warrant interest, and ossibly to stimulate investigat,ion
into other .storms of t 1 is type to discover if some such em irical rule can be laid down; €or if it can, it c~rtt~inly
Let us consider the relation (if any esists) between the distance between the wind-shift line and the 33O isotherm, and (1) the width of the glaze belt 4' of longitude east
For (1) with 34 cases:
For (2) with 28 cases:
For (3) with 28 cases:
\oil P be of value in forecasting.
12 hours later, and (2) the width of the glaze belt 8' of long-itude east 24 hours later. Since the pressure for- mations, and hence the weather in general, move east- ward at a. rate which is about 8' per 24 hours, such a relation would be of value, for, knowing the current dis- tance between the wind shift line and the 3 2 O i..othem, it might be possilde to predict the area over which glaze will form 4' and 8' farther east 12 and 24 hours later, respectively. Such mcasurements were made from the maps with the following result:
Width of g t z e belt ?O east 13 houre later. = 0.9 (4'i:urrent distance between wind-shift line and 3 2 O isotherm.
Width of glaze belt So eaet 24 hours later.
(')C=naistance between wind-shift line and 3 2 O isot.herm. =
Stating the accurac.y in the same manner as in the
For (4) with 39 cases:
previous case we have :
13 agreed within 80 per cent of the mean.
26 agreed within 40 per cent of the mean.
13 agreed within 80 per cent of the mean. 17 agreed within 40 per cent of the mean.
'33 agreed within 60 per cent of the mean.
"he accuracy of the first in this determination is tho greatest. of the whole set, and, if given any relative weight, would he considered the most valuable of the fivc relat,ions presented in the paper. The second is not so reliahle, as might 1 )~ espected, and ranks ahout equal with thosc of the preceding set.
For (5) with 37 c.ases:
COXCLUSION.
The n.wd for .inore o.erohgicul clatcr..-l'he conclusion of
an invest.igation of t,his type leaves much to he desired.
for t.he search for threads amon- the multiplicity of atmospheric phenomena, which, iP traced, may lead to other conclusions of value, is indeed an alluring but. end- less effort. If fifty or a hundred such storms could be followed t,hrough, each supported bv ade uate observa- tions in the uppm air, them is little dou B t that much could be learned which the forecasters could use with siiccess. But each storm is a roblem in itself which e n t d s weeks of laborious t.ab Ig atioii and stud?, and, &o, only those which have occurred within the ast fivu years have even a few data from the upper air. One of the facts which this paper has brought out, however, is that we do not kave enmiggh obsewaths of t h ~ u FT air
b y kites. There are sis kite stations m a i i i t a i n e ~~g the Weather Bureau in the etistern half of the United States. There should be at, least! as mail more.
wind, the vertical tem erature radient., the moisture
them for a large number of points t,hroughout tvhe east- ern half of the count.ry, the forecasting of sleet and glaze-now one of the most uncertain and dificult- would be rendered relatively easy. It, is needless to point out the economy of such a venture, when it is common knowledge that millions of dollars are lost or wasted through lack of preparation for these destructive emergencies. And without the uppcr air data, the fore- casting must rest upon an unst-able empirical basis.
g permitted t,o -know the dept,h o P the under-running nort
content +tad strengt'h o P the sout a erly wind, and know
If we were on1
80 MONTHLY WEATHER R.EVIEW. FEBRUARY, 1920
The naessity for carfful cbud observations.-The value of careful cloud observations can not be overemphasized; carelessly made observations of the kind, amount, and direction of clouds are worse than none a t all, yet the average observer is likely to be less careful in this than in any part of his work. On the afternoon of February 9 a maid of the consulate
tell a great deal when observations by kites and ran into the house gleefully showing a handful of. snow balloons are wanting; but where the latter are a\-*g$i which she had 1)ressed into snowball. It wa,s the first the clouds form useful and necessary supplementary data. snow she had ever seen. AS the weather had been Th.e W;..E-sh$t Zine.-It seem worth wMe to mention constantly inclement since the middle of Novemlmr,
again the importance of the wind-shift h i e as a factor 1919. I supposed that this W ~S but a snow flurry ending
in forecasting. Bjerknes has found it of extreme im or- the former rains, as over 27 inches of rain had already tance in forecasting precipitation; C1 ton: also, a i , fallen, this quantity being far above tlie average at this similar lines, shows the importance ' o lines of conver- time of the pear. In 36 hours afterwards, however.
gence of winds as factors in forecasting recipitation, and Jerusalem and the surrounding country for miles had this paper shows with clearness that t % e wmd-shift lint been mantled by t i snowfall which averaged on the plays no small part iu the production of &et and glaze. level 40 inches, with drifts in many places reaching a liei lit of 10 feet.
ACKNOWLEDQMENTS. I% the memory of the inhabitants of Jerusalem this was the greatest snowfall and the people were absolutely appallttd by it. eat blizzard of 1888 and
States, but none compared in possibilities of danger to this one. The locust visitation of 1915 falls into insig- nificance AS conip::red with it. The people of this countr! lwing unusecl to such n phenomenon were totally unprepared to contend with it,. There were no snow
1km-S 01' even snow- s1iowls, and if there had heen the po1)ulatioii would mt. 1in.w knuwn how to handle them. 811 communication withiu the snow limit was inter- rupted, and the falling of the telegraph wires, the blocking
of the railroad and all t,horoughfares cut us off entirely from the outside world. Every store wns closed. The
DEMORALIZATION OF TRAFFIC IN NEW YORK CITY BY SNOW AND SLEET. There was a shortage of bread and a dearth of wood and kerosene, and starvation and freezing faced the The meteorological record of Fehruary, 1920. at New York. nil1 low be remembered by reamn of the remmkal,le abm-heavy pre- People- Fortunately, there is a battalion of Yorhhire
cipitation which be an to fall in the earl morning of the 4th instant,. troops garrisoning the city. This battalion saved the
and kept it up unti? t.he early morning oythe 5th. During that perind situation. At once over 700 men were a t work with of ahout 75 hours. 4.45 inches of precipitstion occurred. Of thiA shorcls opening roads sild streets the city and amount 0.37 inch m-aa rain; 2.11 inch- of melted sleet; and 1.9;
15.5 inches, of which 8.8 inches were sleet and 8.7 inches [were1 snow. were opclled the spirit of profiteering which was already remorselessly abroad in this community-causing the prices of all necessaries, not to of lusuries, to
Of the New Weather which caused increase from five to t.en t,imes their former value, a t'ie-up Of traffic in the city 1i.aving niade .Jerusalem iii the past year possiblv the more excessive On Ot The most expensive place in tlie world, the cost of living 8CknEi;fie American of Februa being twice as high as in t and in syria-knew no
and the military uovernor was compelled to strenuously streets of tile immense banks of snow or of their foot- restraint. In consequence, mo violence was imminent
deep content. Trucks were to get reduce tho price 07 bread and other food commodities. At least 40 houses in Jerusalem were crushed in by about with facility, traffic ways, only wide enough for a
great tie-u ' efforts t'O pass; flame *Owemf a list is compa.ratively short. When at last unfetterel
the iiiliahitaiits in general proved equal to the occasion device, but these Were SO 10Cd their own provisiolls with the poor and were impractical; the Clonditions are now noma], the telqpph and railroad lines being in operation, stores opened, and The snow structed snow digger was tried with great success T I ut it success, where the drifts were deep enough; a new1
was impossible to get enough of this t p e of machine to native has rapi f; ly disappeared, and escept for the continuedhigh be Of use in the emer ency; the 're hose w88 'Ised m-inds and unusua.lly cold weather one would not believe
I>lizzard will go down to history as one of the most
This all large cities where such storms are a possibility, and
of the Holy City, a city by no means unacquainted emergency.-0. L. M.
with estrsordinary incidents.
TREMENDOUS SNOWSTORM IN PALESTINE, FEBRUARY 9-11, 1920.
By OTIB A. GLAZEBBOOK, American Consul.
IJeriwalem, Palesthe, Feb. 23,19ao.]
Dotailed cloud observations
To Dr. C. E'. Brooks, who riot only suggested the in-
ing at all times to give advice and valuable sug estionu; to Mr. Willis R. Gregg, who kindly laced d t& avail- able aerological data at my disposa P and made helpful suggestions; to Mr. Herbert Lyman, who assisted iii the laborious tabulation of data hom the original records,
which f0i-m the bmis for the accom nnJ-ing &arts; and to others, who, from time to time, Bspfa .ed an encour- aging interest in the study, I most ratefu y acknow-ledgn my indebtedness and express my t anks.
I recall the
vestigation and outlined the methods, but also was will- remember many other heavy Y a118 of snow in tlie United
li
a
Felahin Coulcl not bring their products to tho markets.
inch- c,f melt& RnClw. The maximum depth of Enow anrl sleet wap
noted in Form 1001
in spite Of
28, 1920, pages 219, 232,
dkgug Out the buried POPulrttiO1l. When
This is the record of the storm,
lil
% and 233, gives an account of t T e methods used to rid the
sing1e stream Of but this in the weight of the snow, but, strailgo to say, the casualt
P the warp were tried, another their reSU1tS that they hg
was tried, with hprovidefit.
roducts coming into the market.
with the greatest e d This experience *Odd that the city had 'ust raised itself from the dead.
lead to the Provision Of adequate for remarktlble and dangerous dccurrenc- in tlie &tory
I
- . -
7 Note in thla REVIEW, p. 83.