Hourly and Daily Meteorological Characteristics of the 1957 Shurin

Hourly and Daily Meteorological Characteristics of the 1957 Shurin

Jo㎞C。 KIMURA

ABSTRACT

    Average daily values illustrate that on the Pacific side， more rain falls in central Japan during the Shurin than either in the north or in the south． On the Japan Sea side， rainfall decreases northward・ During the 1957season， there were more storms in northern Japan than elsewhere，

but the duration and the interval betweell storms were shorter． Average duration of precipitation periods was longest in central and sout血western Japan． Precipitation during this season was not intense・

    The begiming of the Shurin seems to be marked by an appreciable decline in both the minimum and maximum temperatures． There was an abrupt drop in the minimum temperature at the end of the season；how−

ever， this was not well defined north of the Kanto Plain． During periods of precipitation， temperature was not only uniform but was substantially affected by wind direction． The magnitude of atmospheric pressure in−

creased as the season progressed． The greatest variation of pressure was found in the northern areas． In the north， average mean relative humidity decreased from summer to the Shurin， while the central and southwestern areas recorded higher figures at the onset of the season．

Average minimum relative humidity indicated a marked decrease in the、

north， but the drop was minima1  in central and southwestern Japan from summer to the Shurin。

    Wind velocity increased from summer to the Shurin． As the storm systems apProached northeastern Japan， they became extremely deep・and high wind velocities were recorded．

ACKNOWLEDGEMENTS

    Appreciation is expressed to Dean Taiji Yazawa of the Faculty of Science and the Department of Geography， Tokyo Metropolitan University，

for his guidance throughout this project． This report would not have been possible without the assistance of the Statistical Section of the Japan Me−

teorological Agency that provided unpublished hourly data． Gratitude is also expressed to individuals and agencies that aided in the completion of this report．

一105一

1 INTRODUCTION

    Precipitation during血e Japanese shurin season is associated with depressions t血at develop on the Polar Front in the East China Sea and progress northeastWard・and those that originate in Siberia a皿d swing in all arc−like manner southeastward， then pilogress．to the northeast1．

Weather charts indicate that， on occasion， typhoons also play an important role in the weather picture during this season．

    A・eries・f rep・rt・2 c・ncerning the Shurin ha・been c・nducted． Thi、

is the third of such reports・ For the period 1957 through 1961， the l957 season was found to be the most representative3， A ccordingly， a des。

cription of the hourly and daily meteorological elements for the Shurin of 1957will be utilized to characterize the season． Daily data for 18 sta−

ti・n・4 and・even h・urly mete・r・1・9ica1 element・f・r eight・tati・n・5 were graphed and analysed for the months of Augtlst， September and October，

1957，and compared with the daily synoptic surface charts．   ．．

    This study was undertakell with the hope of determining characteris−

tic features， patterns and tendencies relative to the amount and intensity of precipitation；frequency・duration and interval between periods of pre−

cipitation；cloud cover；temperature characteristics and tendencies；at−

mospheric pressure variations and tendencies；relative humidity；and wind velocity and direction during the Shurin．

I

I ANALYSIS OF DAILY AND HOURLY METEOROLOGICAL DATA

1． Precipitation

    From the first of August to the beginning of the 1957 Shurin， a com−

parison betweell synoptic surface charts and hourly meteorological data shows that．the frequency of long continued and abundant precipitation was rare． In instances when there was an abundance of precipitation， it was due to one of血ree causes：Depressions developing on the Polar Front in the Mid−Yangtze Regi・n， typh・・n・， and・thunder・t・rm・6．

    An analysis of daily synoptic charts shows that precipitation during the Shurin is caused principally by frontal storms and typhoons． The more important of the two are the frontal storms that originate along the Polar Front in the East China Sea． Before the Shurin of 1957， these troughs originated in the Mid−Yangtze． However， surface charts indi−

cate that numerous depressions were formed in the adjacent East China Sea after the season began． Another important source of origin of the depressions is in Siberia． These swing in an arc−1ike trajectory and affect Japan・

    1） Amount and Deviation from Norma1： Table l portrays precipi−

tation data for selected stations in Japan． The first column represents average・ea・・nal precipitati・n7 and出e、sec・nd・・1umn・h・w・the precipi一

一一

@106．一・

1

tati・n figures f・r th・1957 Shurin8． Akita received a greater am・unt。f precipitation than other stations in Japan during the Shurin of l957． The figure was 513 mm， This figure is l62 mm greater than normal． The

l957 figures diminish both to the north and to the south． Nemuro had 321 and Wajima 337 mm respectively、 while Izuhara recorded only l 19 mm．

During the average season， the figure for Nemuro is 196 mm． The 1957 figu re ・therefore・represents a positive deviation． Wajima冒s average Shurin precipitation figure was not available， but the figure for Izuhara shows a negative deviation of 311 mm．

        On the Pacific Coast， there was a negative deviation north of Osaka， while areas to the south had a positive deviation． The figure for Osaka was about norma1， being a mere l millimeter from the average．

The l957 precipitation figure for Osaka was 292 mm， while the normal for the season is 291． Tokyo， on the other hand， recorded 322 mm in 1957，

and has an average seasonal figure of 420 mm． Miyako also had a nega−

tive deviation． The average is 300， while the 1957 Shurin pre：．ipitation figure was 253 mm． Kagoshima has an average of 292 mm but recorded 416mm during the l957 Shurin。

Table 1 Precipitation for selected stations（．1mm．）

Station Average for Season 1957Shurin Deviation

Akita Wajima Izuhara

Nemuro Miyako Tokyo Osaka Kagoshima

3510 4300 1960 3000 4200 2910 2920

5129 3367

1187、

3210 2529 3220 2920 4155

十1619 一3113

十

十十

125 471 980

  10

1235

Source：Japan Meteorological Agency：Average Daily values of Twelve         Meteorological Elements，Tokyo， July，1957， and unpublished         hourly data

    2） Frequency and Average Amount of Precipitation During Periods of Precipitation： Table 2 shows that the greatest frequency of precipita−

tion periods occurred in the northern portions of the country． According to surface charts・this was related to Siberian storms taking two paths：

the northern route across Hokkaido during the beginning of t血e season，

and a track that swings southward in an arc−like manner from mid−season to the end・ After traversing southward， the latter eventually follows a northeast trajectory・affecting all of Japan． The former affects only the northern portions of the Country． These areas were also in且uenced by

一107一

wave cyclones that developed in the East China Sea and progressed north−

eastward．

        Acomparison of the hourly data with the daily sea−1evel charts show that the storms that invaded northern Japan are not only greater in number but shorter in duration（table 2）． The Siberian and East China Sea depressions converged and alternated in progression with the Siberian Migrating Highs and the result appears to be a general acceleration of the systems in northeastern Japan， especially in eastern Hokkaido．

        Hourly data indicate that the average amount of precipitation that fell per storm diminished northward along the Pacific Coast， while there was a northward increase along the Japan Sea Coast（table 2）． In 1957，

Kagoshima received 59 mm per precipitation period， while the figure was drastically reduced to 32 at Osaka． Tokyo曾s figure increased slightly from that of Osaka to 36， but the figure at Miyako was 28 nim． On the Japan Sea Coast， Izuhara recorded 15 mm． This figure increased to 37 at Wajima， and 51 mm at Akita． Nemuro received an average of 27 mm per period of precipitation during the Shurin of 1957．

    3） Duration of Precipitation Periods： According to an old Chinese almanac， Shurin is defined．as an abμndance of rain in autumn that lasts in excess of three days for each period of precipitation9． The 1957 meteor−

010gical data show that such long−continued periods of precipitation were especially rare in northern Japan．      

        The average duration of a precipitation period ranged from 25．7 hours at Nemuro， to 57．1hours at Kagoshima as illustrated in table 2，

The duration apPears to be the longest along the Japan Sea Coast of Tohoku，

and along the Pacific Coast from Kanto to the southwest． The shortest periods were found in eastern Hokkaido and the Inland Sea region to Tsushima， where the average length of each period of precipitation was but 26．3 hours at Izuhara． This is in accordance with the average rain−

fall per period of precipitation．

        As noted above， Hokkaido had the most numerous precipitation periods， but they were relatively short． Sea−level charts indicate that this was due to the greater number of migrating alld alternating highs and lows that traversed the north． The Tohoku Coast of the Japan Sea also had more periods than the remainder of the country． Unlike Nemuro，

however， the duration was of greater length。 The average period of pre−

cipitation at Akita was 47．4 hours． On the other hand， the duration was shorter along the Pacific Coast of Tohoku． The average duration at Miyako was 28．3 hours，while Miyako had only one precipitation period fewer than Akita， possibly because of the mountain range that exists between the two stations． As the Polar Front migrates southeastward，イhe mountains may act as a rainshadow for Miyako that not only reduces the amount of pre−

cipitation， but the duration of the period of precipitation as wel1， More−

over， synoptic charts show that Miyako is out of range of the stationary fronts that in且uence stations in central and southwestern Japan．

        Although there were fewer storms southwest of the Kanto Plain，

一108一

the duration was greater with the exception of the region from the Inland Sea to Tsushima． The average length of a precipitation period was 34．5 hours in Tokyo， and 57．1hours at Kagoshima． The average length at Osaka was 38．2 hours， but diminished to 26．3 at Izuhara． The Kanto Plain southwestward to Kagoshima on the Pacific Coast was influenced by the greatest number of stationary fronts slightly south of the coastline． More−

over， these areas were also affected by wave cyclones that developed along the stationary front that progressed northeastward． Sea−1evel charts show that these depressions influenced areas to the north， as individual storms， but they were often considered to be continuous storms along the more southerly Pacific Coast．

        The region from the Inland Sea to Tsushima is characterized by having one of the lowest annual precipitation figures in Japan． Meteoro−

10gical data indicate that during the Shurin this area receives significantly less precipitation than other parts of the cguntry except Hokkaido． Syn−

optic charts for 1957 disclose that this area was affected only by the fringe of many fronts that traversed Japan． In addition， there is the rain−

shadow effect of the mountains of southwest Honshu， Kyushu and Shikoku that reduces precipitation figtlres associated with the Polar Front from the northwest and・that associated with the stationary front to the south−

east． This region， therefore， not only had low precipitation figures but also fewer periods of precipitation．

Table 2 Date concerning periods of precipitation， Shurin，

1957

●

Station

No． of P「ec1P・

periods

Interval be−  Average precip． Average dura−

tween pds．of    per pd． of     tion of pd・of precip．（hrs，） precip．（．1mm．） precip．（hrs．）

  Akita   Wajima   Izuhara   Nemuro   Miyako

♂   Tokyo   Osaka   Kagoshima

09829997

ll6 128．5 135．9 96．7 127．9 135．9 124．O l64．2

513 374 148 268 281 358 324 594

47．4 34．2 26。3 25．7 28．3 34．6 38．2 57．1

Source：Japan Meteorological Agnecy（unpublished hourly meteorological

        data）

    4） Interval Between Periods of Precipitation： Hourly meteological data illustrate that the average interval between periods of precipitation

一一一@1 O 9−一一

was greatest in the southwestern sections of Japan． The average interval at Kagoshima was l64 hours，while the figure was reduced northward．

The figure was l36 hours at Izuhara and Tokyo， but decreased to 129 at Wajima，128 at Miyako， and l 16 at Akita． Because of a greater number of storms that influenced Nemuro， the average interval was only 97 hours．

        Although there were slight variations，10ngest intervals between periods of precipitation for most of the stations occurred after the middle of the season． At Nemuro， this was not the case． Storm intervals did not appear to conform to a pattern， and there were several periods of great length．

        DUring mid−season， intervals between storms were shorter than ei由er at the begtnning or the end． Synoptic charts show that the Polar Front extended from south of the Japanese Islands to the mainland of China

°ver the Ea・t China Sea・Depressi・n wave・apPear t・have・riginated・n the fr・nt in the Ea・t China Sea and pr・gressed n・rthea・即ard frequently．

A・these center・m・ved・ver Japan， there were・nly・h・rt peri・d・wh。n the weather was clear．

    5）Inten・ity・H・urly・mete・r・1・gical data indicate that precipitati・n during the Shurin cannot be considered intense。 During the major portion

°fany peri・d・f precipitati・n・the figure wa・1ess than 2 mm per h・ur a，

re且ected in table 3， Moreover， the number of hours that exceeded 6 mm of precipitation was rare。

        Tabulation and computation of hourly data illustrate that for most stati°ns・the average inten・ity・f・precipitati・n， c・nsidering the entire pe−

riod of precipitation， was slightly over l mm per hour． However， the figure・f・r O・aka and Izuhara were well b・1・w thi・average． They were 685and．57 mm respectively． If only those hours with a trace to recorded precipitation are considered the figures increased markedly． Neverthe−

1ess・precipitati・n during this sea・・n i・n・t intense． A・in the previ・u・

example， Osaka and Izuhara had the lowest figures，while the Pacific coast north of Tokyo seems to be in an intermediate position．（table 3）．

        The percentage of hours representing Precipitation of less than 2mm is shown in table 3． If all of the hours from the beginning of the storm， irrespective of those that received no precipitation， are considered，

approximately 84％represented less than 2 mm． Th6 figure for Izuhara，

however・was 93・3％． If only those hours that recorded from trace to 2 mm were involved， the percentages would change drastically． For most stations， the percentage would fall in the range between 45 to 65． The highest figures are for stations northward from the Kanto along the Pacific

coast of Tohoku， indicating that these regions received fewer hours of no rainfall during the average storm．

        Intensity of over 6 mm of rainfall was more frequent in areas northward from central Japan than in the southwestern portions of the country． The percentages were 3．6 at Wajima and 3．9 at Tokyo， increas−

ing to 4・9％at Akita and 3・5％at Miyako， but diminishing to 2．9％at Nemuro．

Intense precipitation appears to diminish in the southwest． At Izuhara，

一110一

2．4％of the storm had an intensity greater than 6 mm per hour， while that for Osaka was 2．0％． Kagoshima  s figure of 3．4％could be misleading，

because the typhoon that passed over the city on September 6 was also in−

cluded in the computations． Precipitation exceeded 6 millimeters for seven of the fifty−seven hours of the storm， but only 14 hours surpassed 6mm of precipitation for the entire season．

        Acomparison between hourly data and synoptic charts illustrates that the most intense precipitation is associated with either typhoon rain。

fall or with the passage of a weather front． Accordingly， intense rainfall is associated with rapidly−falling pressure and at the lowest pressure reading． Intense precipitation is also associated with thunderstorms．

During the 1957 Shurin， there were only a few．

       Table 3  1ntens ity of precipitation

Millimeters of precipitation ^（hours） Aver， Aver．

Total intensity intens ity Station

0 T−2 2−3 3−4 4．5 5−6 6＋ hours _per storm during

（mm，） precip＊

7

Akita 178 220 30 9 9 5 23 474 1．08 1．73

Percentages 36．6 46．4 6．6 1．9 1．9

4．9

Wajima ^{ll9 140 18 9 5 7 11} 309 ^{1．09 1．77}

Percentages 38．6 45．7 5．8 2．9 1．6 2．3 3．6

Iz血ara ⁹¹ 105 5 3 1

5 210 ^．57 1．00

Percentage寧 ^{43．3 50．0 2．4 1．4 0．5} 一 2．4

Nemuro ^{93 165 19 12 5 5 9} 308 ^{1．04 1．49}

Perce夏tages ^{30．2 53．6 6．1 3．9 1．6 1．6 2．9}

Miyako ^{82 143 ll 8 3} 一 9 256 1．00 1．46

Percentages 32．2 56．1 4．3 3．1 1．2 一 3．5

Tokyo ⁵⁷ 209 13 14 4 1 12 310 1．04 1．27

Percentages 18．3 67．2 4．1 4．4 L3 ^{0．3 3．9}

Osaka 73 228 20 10 3 3 7 344 ．85 1．08

Percentages 21．2 66．3 5．8 2．9 0．9 0．9 2．0

Kagoshima 131 209 26 13 7 7 14 407 LO2 ^1．50

Percentages 322 ^{51．4 6．4 3．2}

^3．4

＊ Excluding column 1， hours with no precipitation

Source：Japan Meteorological Agency（unpublished hourly meteorological date）

    2． Cloud Cover

    The overall cloud cover pattern that is re且ected by the hourly and daily data for l957 shows a great similarity to that of the average daily value・10． M・st・f Japan・・uth・f H・kkaid・・n the Pacific b・rderland・and southwest of Tsuruga on the Japan Sea Coast experienced generally clear conditions during the summer． However， the sky became overcast as the season began， and the sky was completely overcast several hours or even several days before precipitation actually began to fall。 Usually clearing occurred several hours after the rain ceased．

一一

P11−一

    There were numerous occasions when the sky was clear during the Shurin． A comparison between hourly data and sea−1evel charts shows that t血e sky was especially clear when the Siberian high pressure cell was in the process of migrating to the North Pacific． Near the beginning and during mid−season， clear skies are visible， but for short periods when compared with periods of overcast condition． Toward the end of the sea−

son， clear periods increased in length as cold fronts moved rapidly south of Japan before the intense advancing migrating highs・

    Cloud cover increased along the Hokuriku and the Japan Sea Coast of Tohoku from summer to the Shurin． It did not decrease to the extent that it did in other parts of Japan， nor was there a decreasing tendency toward the end of the season． After a slight decrease at the end of the season，

the sky once again became overcast as the cool season precipitation maxi−

mum was about to begin．

    On the other hand， in southeastern and eastern Hokkaido， a high per−

centage of cloudiness was recorded during the summer， This tended to、

merge with that of the Shurin． In the remainder of Japan， except for the Japan Sea coast， a marked decrease was noted at the end of the season・

3． Temperature

    1） Temperature Characteristics at the Beginning of the Shurin：The l957 hourly data show that as the Shurin season began， maximum tempera−

ture decreased abruptly． This decline appeared to take place from north to south， a皿d it was first evident in Hokkaido． Sapporo was affected on August 25， after which the effect was felt to the south， then southwestward to Kyushu． Kagoshima recorded a dip on September 2． This decline occurred on similar dates along both the Pacific and the Japan Sea coasts of northern Honshu at correspondtng latitudes．

        Acomparison between hourly data and sea−1evel charts shows that the first major break in maximum temperature began with the first invasion of the migrating high pressure ridge from Siberia． The average minimum temperatures for a period of five days before the abrupt drop were 31．Oo，31，50，31．30，31．20 and 31．50C． at Miyako， Wajima， Tokyo，。ip Osaka and Kagoshima respectively． On August 25， the maximum tem−

perature dropped to 26．50C． at Miyako， while， on the same day， the read−

ing was 29．40 at Wajima． The figure at Tokyo on August 29 was 27，40．

The decrease in maximum temperature occurred on September l at Osaka with a reading of 28．30C． On the following day， Kagoshima recorded 30．40C． Coincident with the reduction in temperature， the sky was clear，

the minimum relative humidity decreased below 40％at many of the sta−

tions， and the wind shifted to the northern component． Subsequent days were similar in characteristics．

        飾

        Anoticeable reduction in minimum temperature took place im−

mediately before the maximum temperature decline． However， average daily values disclose that minimum temperature declines were negligible at the beginning of the seasonl1． Nevertheless， there was a similar pro−

gression from north to south and southwest that was noted for the decline

      −112一

of maximum temperature． The average minimum temperatures for a

peri・d・f five day・preceding the dr・p were 23．9°，24。9°24．9°， and 23．4°C．

at Miyako， Tokyo， Osaka and Kagoshima respectively． On the morning of August 24， Miyako had a low of 22．90， while Tokyo to be south recorded 21．10C． on August 29． On August 30 the minimum temperature at Osaka was l7．60， and Kagoshima recorded 18．90C． on September 2． The tem−

peratures at Wajima were affected by rainfall and no attempt was made to ascertain the minimum figures ． Nemuro雪s temperatures did not conform to the normal pattern， hence no attempt to determine the decrease was made．

        The foregoing shows that the decrease in maximum temperature was more marked in the northern than in the southwestern areas， Synoptic charts show that while the northern portions of Japan were dominated by Siberian air， the southwestern sections were also under the in且uence of the Ogasawara High． The perplexing problem concerns the greater de−

crease in minimum temperature figures from north to south， or the rever−

sal of the trend of maximum temperature decline． Analysis of surface charts did not reveal an adequate explanation for this phenomenon． How−

ever， daily meteorological data indicated that during the early part of the Shurin， the sky was clear when the Siberian air was dominating Japan and the daily march of temperature remained unaffected， As a result， the decrease in minimum temperatures is greater than the decline of maxi−

mum temperatures・

        Although both the maximum and minimuM temperatures became lower， there was a wide temperature range． The figure was between IOo to l30C． For example， at Kagoshima， the average temperature range for aperiod of five days preceding the abrupt maximum temperature decrease wa，8．7・C． On September 2， the t・mperature range increased t・11・5°C・

As the season progressed and overcast conditions became widespread，

temperature rang・decreased t・apPr・ximately 7°t・8°C． f・r m・・t・f Japan． The rang・f・r Nemur・wa・・nly ab・ut 5°C・

    2） Temperature Characteristics During Periods of Precipitation：

During periods of precipitation， temperature is characterized as being ex−

tremely uniform despite the diurnal heating and cooling processes． The range of temperature was less than 30C． at most places because of air mass homogeniety and overcast sky conditions． During the seventy−hour period of precipitation that resulted from a stationary front on September

9，the temperature range at Kagoshima was 3．60， while at Osaka and Tokyo the ra阜ge was 4．70 and 3，70 respectively． The variation in temperature during aごold front passage that resulted in precipitation on September 25 was 3。lo at Wajima， and the difference between the maximum and mini−

mum for that day at Miyako was 2．30，while・Nemuro recorded a difference

ofamereO．80C．

        It has been noted that temperature will be uniformly high during periods of precipitation if there is a southerly wind component． Average maximum and average minimum temperatures were determined from hourly data for a period of five days before and an equal number of days

e

●

L−113一

afte「the passag・・f a c・ld fr・nt・n S・pt・mber 25， f・r Miyak・． The

average maximum temperature was l9．50C．，and the average minimum temperature wa・1・・7°・The average t・mperature fr・m th・beginning t。

the end of the period of precipitation was l6．80C，，or 2．70C． lower than the averag・maximum and 5・1°C・hlgher than the average minimum t。m−

perature．

        On the other hand， temperature is uniformly low if the wind is f「°mth・n・rthern・・mp・nent・Similar calculati・n・were c・mput・d f。r T°ky°a・in止e preceding Paragraph・There exi・ted a pred・minantly north component wind during the precipitation that resulted from a sta−

ti°nary fr・nt that was situated・lightly t・the s・uth・f th・Japanese l。land。．

period of five days preceding and a like number of days after the storm．

     3） Temperature Characteristics at the End of the Shurin： The ab−

「upt and marked decline in temperature・e・pecially in minimum t・mpera−

 ture， was not well defined north of the Kanto Plain on the Pacific side；

 north of Tsuruga on the Japan Sea borderland and in Hokkaido， because there existed a number・f decline・． S・uthwest・f these area、， there was amarked drop in both the maximum and minimum temperatures at the end

°fthe・ea・・n・On O・t・berユ7・a・a・tr・ng migrating high pressure cell became・ituated・ver Japan・b・血th・maximum and minim・m temperat。reS declined rapidly and abr・ptly血r・ugh・ut the entire c・untry．

         T°ky・・0・aka and Kag・・hima are pre・ented a・example illu，trat−

ing山e declin・・f b・th・maximum and・minim・m t・mperature・． Averag。，

we「e c°mputed fr・m h・urly data f・r a peri・d・f five day・b・f・re the ab．

「upt tempe「atu「e dr・P・A・・mpari・・n with similar c・mputati・ns after the decline indicate・the f・11・wing・At T・ky・， the averag・maximum tem．

3：1艦翻慧a鷲￡tf蹟窪器も器濃 ；瀦邑r蹴濫

窪4鷲膿器搬溜鷲さど・f。鵠賭翻。鼎灘．

盈膿臨：器：S。yfgl，9 言：，9；：糠朧邑n識臨編V購．

after the dr・P・0・aka ・figure・were 14．6°and 9．4・b。f。re and after th。

decline re・pectively・while山・・e f・r Kag・・hima were 14．2・and 10．6・C．

The ab・ve figures indi・at・that there wa・agreater declin。 in b。th maxi．

mum and minim・m t・mperat・res in血e n・rthern p・rti・n， than in th。

−southern areas of the country．

￡灘搬u㍊e灘孟畿濫二粥：認：le翻農雛欝・f

significant p°int・f demarcati・n at the end・f血e・ea・・n， the・957 h・urly

一114一 ^●

data disclose a significant drop in the centra⊥and southwestern areas．

Synoptic charts show that this was related to an influx of Siberian air．

        There are numerous instances of marked reduction of both maxi−

mum and minimum temperatures north of the Kanto Plain on the Pacific side， and north of Tsuruga on the Japan Sea side． Therefore， it is diffi−

cult to determine the end of the Shurin for an individual year by tempera−

ture characteristics alone．

4． Atmospheric Pressure

    1） Pressure Variations During the Shurin： Surface charts illustrate that summer pressure pattern persisted during the early part of the sea−

son southwest of Osaka， and variations were noted to be minor。 Hourly meteorological data illustrate that the difference between the depression that passed Osaka on August 31 and the maximum reading of the ridge that followed was 8．1mb．、At Hiroshima， the variation was only 7。O mb．

Similarly， a maximum reading of lOl6．8mb was recorded on September l4， at Kagoshima， and two days later a minimum of lOO9．9was recorded．

Excepting typhoons and influences of typhoons，these areas reflected only asmall fange of atmospheric pressure during the early stages of the Shurin． In northern Japan， there was a greater difference between mini−

mum a耳d maximum pressure readings． The difference between the mini−

mum figures ascertained from a trough that passed on August 31 and the maximum readings obtained from a ridge on September 5 show that the greatest variation existed in Hokkaido because of the deepening of depres−

sions as they progress northeastward． Tokyo had a low reading of lOOl．6 and a high of lOl7．2 mb． Miyako  s difference was slightly greater than the figure at Tokyo， recording lOOO．7 and．1017．9mb for the minimum and maximum pressures respectively， while Wajima had readings of lOOI．7 and lO15．4 mb． On Hokkaido， the variation increased greatly． Sapporo had a minimum pressure on August 310f 998．2， and a maximum of 1015．4 mb on September 5， while the difference is even more striking at Nqmuro，

where the pressures were 995．O and lO17．7 mb．

        During mid−season， southwestern Japan still experienced o亘ly a slight change in pressure readings from minimum to maxir曲m figures．

However， a noticeable increase was evident north of the Kanto Plain．

Sea−1evel charts show that as the season progressed， the high pressure cells became stronger and the troughs deepened． In the southwest， the high pressure cells increased in strength but the depressions did not deepen appreciably． As a result， atmospheric pressure range was lower than in other parts of Japan． By the end of the Shurin， even the south−

western areas show a great variation． This is no way compared to the variation that existed in northern Japah．

    2） Pressure Tendencies： Atmospheric pressure gradually rises as migrating highs invade Japan． By comparison， pressure readings fall ab−

ruptly as a trough approaches． Daily meteorological data indicate that

一115一

increases occur over a period o工several days，but a decline will follow within ten to fifteen hours after the ridge passes． For example， pressure gradually rose for seven conSecutive days from September l8， but all stations experienced a rapid decline dユat lasted a maximum of fifteen hours as a depression apProached．

5． Relative Humidity

    Table 4 shows that throughout the summer weeks the average mean daily relative humidity was constant and high． The figures ranged from 75％at Osaka， to slightly over 80％for most of northern Japan， except Nemuro， where the figure was 93％二．Southwest of Osaka， the average mean daily relative humidity was slightly higher during the Shurin than it was during the summer weeks，while the pre−season figures were higher in areas north of Osaka． In the southwestern portions of the country， the average mean daily figures during the Shurin were slightly higher because those areas were under greater influence of the moist trough conditions and the Ogasawara High than were the areas to the north． Synoptic charts reveal that it is probable that the average mean relative humidity figures

      ダ were lower in areas to the north because of the increased activity of the migrating high pressure cells that originated in Siberia and the reduced in且uence of the Ogasawara High．

冒 Although the average mean dailY r61ative humidity figures ctecreased from summer to the Shurin in areas north of Osaka， there were a greater number of days during which the mean daily values were higher than they were for days in the summer． This suggests that the minimum figures during the in且uence of the migrating high are extremely low， while during most of the season they are affected by moist conditions．

    The foregoing is also reflected in the average minimum daily relative humidity figures． During most of the season， the minimum daily values were greater than the figures for the period before the Shurin． However，

as the Siberian High invaded Japan， minimum daily figures declined to a low of 35％． These occasions contribute to the slightly lower average mean р≠奄撃凵@relative humidity values during the Shurin than during the summer． Only southern Kyushu had average minimum daily figures dur−

ing the Shurin that were greater than the summer figures． The figure at Kagoshima was higher only by one percent． Hiroshima and Osaka show that the average daily minimum relative humidity was 2％lower during the Shurin than from the period August l to the beginning of the Shurin． But to the north， this figし1re increqsed to slightly over lO％・

    Most df Japan shows a great decrease in average mean and average minimum daily relative humidity from the end of the Shurin to the end of ． October． Although the figures decline， the decreases are lower than ex−

pected along the Pacific and Japan sea coasts northward from central Honshu except northeastern Hokkaido． The problem is not well under−

stood along the Pacific coast， because the end of the Shurin marks the beginning of the autumn and winter dry s easons． On the other hand， along the Japan Sea coast of Hokuriku and Tohoku， the end of the Shurin season

一116一

Table 4 R。1ativ。 humidity ≠獅п@wind∀・1・city，・1957

 Average rnean relative humidity

      （％）

Average minimum

relative humidity

      （％）

Nemuro

 Atlg． 1−Aug．23

 Aug．24−Oct． 9（Shurin）＊

 Oct． 10−Oct． 31

 Aug． 1−Aug．23

 Aug。24−Oct． 9（Shurin）

 Oct． 10−Oct． 31  Aug． 1−Aug．23

 Aug．24−Oct． 9（Shurin）

 Oct． 10−Oct． 31  Aug． 1−Aug。23

 Aug．24−Oct． 9（Shurin）

 Oct， 10 − Oct． 31

 Aug． 1−Aug．31

 Sept． 1−Oct． 17（Shurin）

 Oct． 18 −Oct． 31

0saka

 Aug． 1−Sept・ l

  Sept． 2−Oct． 17（Shurin）

  Oct， r8 −Oct． 31

Hiroshima

  Aug． 1−Sept・ 4

  Sept。 5−Oct． 17（Shurin）

  Oct．  18 − Oct． 31

  Aug． 1−Sept． 3

  Sept． 4−Oct． 17（Shurin＞

  Oct． 18 −Oct． 31

・Date、。f・the・Shurin Cea・・n・f l95ラ， a・determined by daily data．

Data from Monthly Report of the Japan Meteorological Agency，

August， and October，1957， and unpublished hourly data．

      −117一

Average wind

  velocity    （mps）

988  498 345  322

424

447

門ノQ110

Q／Q／4

の

、

3〆0ー

の

means that the winter maximum Qf precipitation is about to begin．

6． Wind Velocity

     The average wind velocities were computed from hour1シdata and are presented in table 4． They do not include winds during typhoons， tropical storms or depressions resulting from the downgrading of typhoons be−

cause of their overwhelming effect in marring the averages．

     Average daily wind velocities are lower from August 1， to the begin−

ning of the Shurin than during the Shurin season． Synoptic charts show『

that the summer isobaric pattern over east Asia is relatively uniform，

With the onset of the season， increased storln activities with deep troughs are evident． Accordingly， wind velocities should increase． There was also an increasing trend after the end of the season． There al e， however，

stations that did not conform to this pattern． At Osaka， the wind did not increase with the beginning of the  season， but remained relatively stable，

and the average daily wind at the end of the season decreased in velocity．

Tokyo had an average wind velocity that decreased from summer to the Shurin， and the increase that is noted for the end of the season was lower than the summer figures． In addition， Sapporo had a decreasing wind velocity tendency， but Nemuro recorded an increasing tendency， with highest wind velocities of aUJapan during these three periods．

    H・urly data and・）m・ptic charts馳illu・trat・t五at『the highe・t ve1・city wind・are c・incident with the passage・f a center・f l・w pressure・r a c°ld fr・nt・During this sea・・n・wind ve1・citie・increa・ed t・bey・nd 10 mps with the passage・f tr・ugh・and fr・nt・・M・re・ver， in many in・tance・

the figure exceeded 20 mp・at Nemur・・a・depressi・n・became extrem・1y deep as they progressed to northeastern Japan．

7． Wind Direction

    H・urly・mete・r・1・gi・al data・h・w that bef・re th・beginning・f the 1957 Shurin・wind directi・n in m・・t・f Japan fav・red the・・uthern・c・mp・nent．

The greatest frequency along the Pacific coast was from the southern direction；along the Japan Sea coast and in the Inland Sea， winds favored not only the southwest but the north and northeast as we11． At Wajima，

from August l to the beginning of the Shurin， the wind was from the north−

east 49％of the time， and from the southwest 44％． At Osaka， nort㎞orth−

easterly wind prevailed 27％of the time and southwesterly 44％． At Hiroshima， there were north and northeast winds 44％and 27％of the time respectively．

      The figures for Hiroshima represent seven consecutive days of north component winds immediately before the Shurin，

    Northern component winds prevailed during most of the season at all stations except along the northern Pacific coast of Tohoku and eastern Hokkaido． Precipitation appears to be strongly associated with these north component winds． At Kagoshima and Osaka， precipitation was associated with northeast winds． At Tokyo， the wind was primarily from the north−

west and secondarily from the northeast． Hiroshima and Wajima show

一118一

both north and nort㎞ortheast winds during periods of precipitation．

Miyako and Nemuro had variable winds while precipitation was occurring．

However， both northwest and south winds appear to be favored at Miyako，

while northeast and southeast wind influences were felt at Nemuro during periods of precipitation．

    North component wind becomes a strong influence  after the end of the Shurin． Miyako and Nemuro did not appear to conform to this pattern immediately after the end of the season．

CONCLUSION

    Daily data for August， September and October，1957 through l961 for eighteen reporting stationS in Japan were compared with the average daily values for these stations， The 1957 Shurin was found to be the most rep−

resentative． However， data for a period of only five years were exam−

ined， and， as such， numerous irregularities wer6 noted．

    Analy・i・and c・rrelati・n・f the 1957 mete・r・1・gi・al data with the surface charts show that most of the rain during the Shurin was associated with depressions that developed either on the Polar Front in the East China Sea， or that which originated in Siberia． Average daily values indicate that on the Pacific side precipitation is heaviest in central Japan and diminishes both to the north and to the south during．not ina1 year． On the Japan Sea side・rainfall figures decrease northward． During the 1957 1 seas°n・h・wever・precipitati・n figure・were 1・wer than n・rmal except at Kagoshima and Nemuro， while on the Japan Sea coast， northern areas received more rainfall than did the southern sections．

    It was found that the northern parts of Japan were affected by a greater number of storms than the remainder of the country because these areas were influenced by both depressions that developed on the Polar Front and in Siberia， while the Siberian storms often traversed the northern track and did not affect the southwest． Typhoons also play a role in the weather picture of Japan． Several varieties of storms affected Japan during the Shurin・An analysis of meteorological data and synoptic charts for a period sufficient to be considered representative to deter−

mine the relative importance of each storm type for specific regions may be of significance． Moreover，中e amounts and percentages of precipita−

tion associated with each type may be regionally compared with．the total average rainfall for the season．

    In conjullction with such a study， the average duration of each storm type may also be calculated for regions． This report combines all storms and illustrates that in the northern areas not only the average durations but the intervals between periods of precipitation were shorter， while in central and southwestern Japan they were longer・

    Data cgncerning aVerage intensity of each type of storm may also be determined from the aforementioned type of investigation． For the year 1957，hourly figures disclose that precipitation during the Shurin was not intense． Less than 2 mm per hour fell during 85％of the time that rain

一119一

was falling， while raihfall measurlng more than 6 mm per hour represented less than 5％of the entire pe籠iod of prΩcipitation． An additional method of determining precipitation intensity might be based on the amount． For example， the amoμnt of intense rainfall as compared to other categories of intensity for individual storms，storm types or the entire season could be calculated．

    Although topography was hot discussed as an important factor in

．influencing Precipitation characteristics・it is n（ミvertheless a factor of concern． Because storms can be traced on synoptic charts， average storm tracks can be determined， and analysis of average and specific meteorological data can indicate the significance of topographical effects as they affect precipitation．

    Daily clbud cover for most stations re且ected similarities with the pattern・f・the average daily・value・・Extreme c1・udiness in・・uthern and eastern Hokkaido during the summer was also noted for the hourly data．

lnls tended to merge with that of the Shurin． Synoptic charts did not disclose an adequate reason， although several weather fronts were噌 observed to haye一婁ravers（…d護he area・        

    At the beginning of the Shurin， minimum temperature declined appre

、ciably． This was followed by a less severe dip in the maximum

figures． These de6iines took place from north to south and occurred on b・th the Japan S・a and th・Pacific c・a・t・at・・rre・p・nding・latitudes at about the same time． At the end of the season， there was an abrupt drop，

e・P．ecially in the minimum figure・・but thi・Wa・n・t w・11 defined n°r血゜f central Japan． N・rth・f the H・kuriku and th・Kant・c・a・t…everal marked declines in minimum temperature were noted． If minimum tem−

perature i・t・be c・n・idered a・・ne・f・the crit．eria・ma・king the end°f the         ，it would be difficult to interpret which decline is significant． season14

    During Peri・d・・f precipitati・n， temperatu・e wa・relativ・ly unif°「m・

It was observed that if the prey皐iling wind was from the southern sector，

temperature wa・high， but if a n・rth・・mp・n・nt wind prevailed・it was extremely low・       ・

    During the early part・f the sea・・n・th・variati・n in atm・・phe「ic pressure was slight in th…uthwe・tern・ecti・n・・f Japan・A・the seas°n pr・gressed， th・magnitud・be・am・g・eater・b・t thi・in n・way c°mpa「ed t・th・great・xtreme・that were rec・rd・d in th・n・rth・Pressu「e in−

crea。ed・1・wly and・teadily a・the Siberian High advanced；h°weve「・1t

declined rapidly as weather fronts proceeded．

    In n。rthern Japan， average mean relativ・humidity decrea・ed f「°m summer to the Shinrin because of the increased influence of the Siberian

High． On th・・ther hand， th…uthw・・tern areas sh・w highe「figu「es at the onset of the season because these areas were being affected by the moist conditions associated with the Shurin． As comp孕red to the northern

，ecti。n，， the effect and frequen・y・f the Siりerian High were reduced in the

。。uthwe，t． The averag・mean and averag・minimum relative・humidity figures sh・w a great decrea・e in th・n・rthern and・・uthweste「n a「eas at  the end of the season， but the drop is negligible along圭he戸okuriku and

一120一

Tohoku coasts of the Japan Sea， and northward from Osaka along the Pacific coast， except in Hokkaido． Because the end of the Shurin marks

         り

the beginning of the drier half of the year along the Pacific coast・the figures should decrease substantially． The slight decrease in relative humidity along the Pacific coast， therefore， remains a mystery．

    There is a general increasing wind velocity from summer to the Shurin and to the end of the season． This trend is considered to be normal be−

cause of the increased storm activity during the season and the greater variation of atmospheric pressure at the end of the Shurin． As storm systems approach northeastern Japan， they become deep． At such times，

extreme wind velocities were noted． However， at Wajima， Tokyo and Sapporo， an actual decrease in the average wind velocity occurred at the onset of the Shurin， and at Osaka and Sapporo the average wind velocity decreased at the end of the season． Synoptic charts did not give a satis−

factory explanation for these problems．

REFERENCES CITED

1

●

2

●

3

コ 4567

8

9；

10．

11．

  6∠Gδ

14．

KIMURA， J． C．：Distinctive weather patterns during the 1957 Shurin．

Science Reports gf the Tohoku University（Geography）， vol．16，1967，

p．41−67．

      ：ibid．，and The beginning and the end of the Shurin season of Japan． GeographicaユReports of the Tokyo Metropolitan University， no・1，1966， P．113−138．

      ： Distinctive weather patterns during the l957 Shurin， OP． cit．

ibid．

ibid．

ibid．

JAPAN METEOROLOGICAL AGENCY：Average Daily values of

Twelve Meteorological Elements， Tokyo， July，1957．

      ：unpublished hourly data for August， September

a耳dOctober，1957．

HATTORI， u． and KOYANAGI， s．：Enlarged Chinese−Japanese Dic−

tionary with Minute Explanations． Fuz anbo，°Tokyo，1941，2031 p．

JAPAN METEoRoLoGlcAL AGENCY：Average Daily Values of

Twelve Meteorological Elements． op． cit．

KIMuRA， J． C。：The beginning and the end of the Shurin s eas on of Japan・OP。 cit．

ibid．

ibid．， and JAPAN METEOROLOGICAL AGENCY：Average Daily

Values of Twelve Meteorological Elements ， op． cit．

KIMuRA，J． C．：The beginning and the end of the Shurin season qf

Japan， OP． cit．

一121 一