トップページ - 横浜国立大学学術情報リポジトリ

全文

(1)Summer High-Level Cyclones in the Region of Japan ''. '. ' By. '. '. Tsugio MASUI (Received Feb. 9, 1962). Abstract In the present study, the development, path and thermal structure of the high-level cyclone which appears in summer and traces a very erratic path in. the upper troposphere over the region of Japan are investigated from the synoptic point of view. Generally speaking, this cyclone is formed by the extreme meandering of the westerlies, but it is not an intense cyclone and has few immediate influence upon the pressure pattern at sea level. In g 1, the author describes briefly the movement of the high-level cyclone in the region of Japan throughout a year. In g2, the characteristics, frequency and path of the summer high-level cyclone are explored and the cyclone is classified into 6 types, including the cyclone in the Bai-u season, according to their paths. The frequency of the cyclone has increased in recent years and its maximum was presented in the year 1958, while the sunspot maximum was reached in the year 1957. But it is not clear for lack of data that there is an intimate relationship between. the extreme meandering of the westerlies in summer and the increase of sunspots.. The path is very erratic compared with that of the high-level cyclones. in other seasons and the movement is steered by the wind in the upper troposphere near the level of the jet stream, or, in other words, by the distribution of anticyclones in that level. Among these cyclones is included the. one which moved perfectly westward from South Japan toward South China in the year 1959. The movement of this cyclone was steered by the easterlies south of the subtropical anticyclone which increased in intensity after the cyclone development. In g3, thermal structure of the cyclone is studied using vertical cross sections. The figure of vertical cross sections is not so noticeable as that of the high-level cyclones in other seasons, but both cyclones belong to the same type except for the existing level. And the structure differs far from that of typhoons in the same season.. In g4, the deepening of the trough in which the cyclone is formed is discussed applying the equation of the vorticity change of a travelling air. t.

(2) 22 T. Masui 1. particle. The term including the vertical velocity is mainly explored and the order of magnitude of it is compared with others. Though the descending velocity'is)estimated at several cm per sec., the order of magnitude of this term is not large, because the wind shear is not large in this season. The trajectory of a descending air parcel is determined using the potential temperature. Descending air parcels can not penetrate into lower layers on. ,. .. account of the high stability in these layers; the cyclone, therefore, exists only in the upper level. Finally in g5, typical examples of the cyclone are examined to ascertain the theoretical process carried in this study.. Introduction In summer, including the Bai-u season, we can see often weak high-level cyclones which follow abnormal courses over the region of Japan. Most of these cyclones develop in the troposphere higher than the 500-mb surface and have directly few relations to the pressure distribution at sea level. During the period after the end of the Bai-u season, the frequency of surface cyclones decreases and subtropical anticyclones predominate over Japan. Particularly, in this season high-level cyclones following very erratic courses develop frequently in the upper troposphere. In this paper, this kind of cyclones is called, for convenience, "summer high-level cyclone". Of course, even in this season we have sometimes intense surface fronts or cyclones and normal. high-level ,cyclones coming from the north, instead of summer high-level. cyclones. ･. The annual frequency of summer high-level cyclones varies widely from. year to year and it is noticed that in recent years the frequency of this sort has increased slightly. It may have some relations to the change of patterns. of general circulation. In fact, we have recently observed the circulation with relatively strong meridional currents over Japan in summer, while the number of sun-spots has increased considerably and the maximum was shown in the year 1958. But we cannot decide whether or not any relation will exist between the frequency of high-level cyclones and the number of sun-spots, in the present circumstances. A high-level cyclone is mainly steered by anticyclones surrounding it or northward-moving typhoons, and consequently its track is very erratic. It i$. believed that a high-level cyclone is formed when the westerly current meanders extremely; its predominant level, therefore, is in the vicinity of the. jet stream. In some cases it may originate in the easterlies, but we have few evidences of this fact., Generally speaking, the development of a summer high-level cyclone has a few or no relations to the lower troposphere; hence the deepening of this cyclone being caused, probably, by the cold advection from the north in the upper troposphere.. .. ,.

(3) , SummerHigh-LevelCyclonesintheRegionofJapan 23 The thermal structure of a sumrner high-level cyclone in the mature stage has a similar feature to that of a high-level cyclone in the cold season except for the difference of the existing levels and it differs far from that of a. typhoon in the same season. In this paper investigation was made chiefly into the movement, thermal structure and development of summer high-level cyclones. The period subjected to the research was confined to JunervAugust, from the year 1952 to. 1959. And among the principal data were included the Upper Air Charts (1954tv1958), Dairy Weather Maps (1958tv1959), Aerological Data of Japan and. Geophysical Magazine published by the Japan Meterological Agency and the Synoptic Weather Maps pubiished by the U.S. Weather Bureau.. g1. HighLevelCyclones. t During all the seasons, we can see, on the upper air charts, many highlevel cyclones coming fro-m Siberia towards the region of Japan and then moving ndrtheastwards. In the cold season an upper trough or a high-level cyclone is formed by the intense outburst of the very cold upper air over the region of Siberia and has the strong tendency of influencing on the development of a cyclone at sea level in advance or in the rear of it. This surface cyclone develops frequently into an intense one in the area to the east of Japan. In winter, the most pronounced level with high-level cyclone is in the middle troposphere [9] and the height of the tropopause in its central portion is often lowered to the 500-mb level or so in the neighbourhood of Japan [5] [7]. Sometimes a high-level cyclone and a surface cyclone in advance of the former are combined into a deep cyclone extending from surface to upper level and its central pressure at sea level reaches the value. lower than 980-mb in the Sea of Okhotsk or Bering. Though the southernmost point on the path at 500-mb level of high-level cyclones oscillates widely in a direction east-west, it is situated generally in. North Japan near 1400E. High-level cyclones pass over this region normally once every 3 or 4 days as is the case with surface cyclones and most frequently when a blocking high persists to the north of Japan. In an extreme case the southernmost point on the path reaches the latitude of about 34eN. For example, on March 24tv25, 1952, a high-level cyclone crossed Central Japan and moved as far to the south as 340N off the Pacific coast of Japan [3] [5].. Towards the warm season, though the frequency of high-level cyclones does not vary appreciably, the structure of them changes gradually and the tropopause height in the central portion rises, too. And their influence on lower layers decreases step by step; hence the development of surface cyclone becoming weaker, except for special cases, for example, the case of the socalled May Storm. In the Bai-u season the weather is subjected to the frontal activity rather.

(4) 24. T. Masui. than the cyclone. In this'season, beside the normal high-level cyclone, we. can see sometimes the high-level cyclones which move southwards in the. ,. region to the west of Japan and then eastwards off the south coast of Japan, and among them exists the cyclone which is not directly related to the surface pressure pattern. This sort of cyclone'is included in the summer high-level. .. cyclone in this paper.. After the Bai-u season the jet stream shifts rapidly farther to the northern latitudes andthe normal high-level cyclone comes to appear far between.. During midsummer only the high-level cyclone having few relations to the surface pressure pattern dominates over Japan. The summer high-level cyclone may develop in the same manner as the normal high-level cyclone in the cold season; namely it is formed by the descending current of the cold air in the vicinity of the jet stream. This subject will be dealt with in g4.. Towards the end of August increases the frequency of normal high-level cyclones again in North Japan and also the frequency of typhoons coming. from the low latitudes to the neighbourhood of Japan. When a typhoon moves near Japan, the jet strem shifts northwards and the development of summer high-level cyclones diminishes in the region of Japan. Particularly in September, we have no summer high-level cyclone in recent years (since the year 1952 and before 1952 it is not clear). In autumn the normal highL level cyclone increases in intensity and the typhoon follows the path to the south of Japan. In this season we could see a few middle layer cyclone de-. veloping only in the middle layer since the year 1957 [6]. Late in autumn we have the first outbreak of very cold air in most cases.. g2. Summer High-Level Cyclones In general, the summer high-level cyclone belongs to the extra-tropical cyclone as mentioned in g1, in summer, however, the stability of lower layers is high in usual cases and consequently the instability of upper layers dose not reach lower layers directly. In such a case we can see a summer highlevel cyclone developing only in the upper troposphere. It is not rare that lower layers are covered with the subtropical anticyclone and the weather is fine when a summer high-level cyclone is migrating in the upper troposphere. This shows the non-existence of an intimate relation between both layers. In the Bai-u season, however, lower layers are instable while a summer highlevel cyclone moves in upper layers, but the instability, both in upper and lower layers, is not closely related at all. In such a case the instability in lower layers is caused by a preceding down current of the cold air.. Summer high-level cyclones develop mainly during the months Junetv August, but intervals found in the course of their developments are not the. same in length. For example, in the Bai-u season, the frequency of them varies widely from year to year. In the year having the predominant Bai-u season or during the period when an intense blocking high persists to the. ,. ..

(5) Summer High-LeVel. Cyclones in the Region of Japan. Table I. List of summer high-level cyclones during the years 1952r-1959.. Year 1952. Date 9.t-11. VI. Jul. 19･-22. V V. Jul.. 21N23' JuL 23N27 Aug. 12N17 Jul.. 1953. 1954. Jul. 10--16. Aug.. 8t--1O. Jul.. 18N24. Central Japan. VI IV. Central Japan. From Oou. Aug. 23.-.-27 Aug. 26-31. III. Western Japan Around Iwojima. 6r-12. Off the east coast of Japan. Aug. IN 7 Aug. 10.-16 Aug. 20.v Aug. 29N31. VI VI VI VI VI VI VI. 12N22. VI. Off the･ east coast of Japan. Aug. 1-- 5 Aug. 12N. II. 3n- 7. Jul.. Jul.. II. VI. v v. Off the east coast of Japan Off the east coast of Japan. (Trough: west-east). 1/ //. 11. tl. ll. tl. From the far south of Japan Off the east coast of Japan Off the south coast of' Japan. May. 31tvJun. 4. Jun.. 21.-.i27. Jul.. 6N11 10N18. VI. Jul. 29r-31. 4N 9. VI VI. 8N 13. II. r/ From the far south of Japan. v. Off the south coast of Japan. II/. I. From the south of 4YV In the neighbourhood of Kadena; northwestward Westward from Japan. I. Far south of Japan. Jul.. Aug. Aug. 1959. III. Off the east coast of Japan Off the east coast of Japan. Jul.. Jul. 24.-27. 1958. From Central Japan In the neighbourhood of Iwojima To the west of Iwojima. VI VI. Aug.. 3.---14. Jul. 14n･v2O. 1957. Off the east coast of Japan. From Oou. Aug. 11r-18. 1956. II. Main area covered by the cyclone. VI VI VI. Jul. 28.-30. 1955. Type. Jun. 26t--Jul, 1 Jul. 13r-19 Jul.. 23N27. Jul.. 27NAug. 4. Aug. 15r-22. v. (IIi/). 1/ Off the east coast of Japan. Off the south coast of Japan Off the east coast of Japan. 25.

(6) T. Masui. 26. Table II. Monthly and annual frequency of summer high-level cyclones during the years 1952･v1959. .. Type INV Year. June. July. Type VI August. Total Total e. 1952. o. 3. 1. 4. 1. 5. 1953. o. 1. 1. 2. o. 2. 1954. o. o. o. -o. 3. 3. 1955. o. o. 2. 2. 2. 4. 1956. o. o. o. o. 7. 7. 1957. o. o. 1. 1. 2. 3. 1958. 2. 1. 1. 4. 3. 7. 1959. 1. 3. 1. 5. o. 5. Total. 3. 8. 7. 18. 18. 36. north of Japan, no summer high-level cyclone develops for lack of conditions. In the late Bai-u season or in the midsummer, in some years, frequently develop large scale troughs to the east of Japan and high-level cyclone migrate. within them. Some of these cyclones may develop into'sUmmer high-level cyclones. During this period, however, no summer high-level cyclone appears over Japan, because in this case the subtropical anticyclone dominates the region above the Japanese Islands even in the upper troposphere. The movement of a typhoon controls the development of a summer high-level cyclone.. When an intense typhoon exist in the neighbourhood of Japan, no summer high-level cyclone appears, but in a special case, when a typhoon contacts directly with a high-level cyclone coming from the north, we have an apparent summer high-level cyclone. The monthly and annual frequency of the summer high-level cyclones is t. illustrated in Table I and II. In the year 1953 and 1954, the frequency of the. cyclones reached its minimum value and in recent years it has increased irregularly, having its maximum in 1958. In 1959, even a westward-moving cyclone appeared, though slightly less frequently. In general, when a summer high-level cyclone eXists near Japan,the jet stream shifts northwards and the. upper wind is not strong and meanders extremely, showing the low zonal index. The same may be said of the whole summer season in a year having many summer high-level cyclones. Though the feature of the jet stream on the whole hemisphere is not investigated, it is clear that the development of. summer high-level cyclones is not a local phenomena in the Far East as seen in the case case of high-level cyclones in the cold season [14]. As shown in Table I, in the year 1954 we had a minimum value of the frequency and in. the same year, the sun-$pots number reac･hed alsQ its minimum value. On the other hand, in the year 1958, both numbers reached their maximum values. There may be some relations between the extreme meandering of the circu-. ..

(7) Summer High-Level Cyclones in the Region of Japan 27 lation and the maximum value of the sun-spots number; there is, however, a fact that in the year 1955 we had a little many summer high-level cyclones and in 1956 a few.. Next we shall consider the path of the summer high-level cyclones, as their thermal structure and development will be discussed in g3 and g4. In the Bai-u season, the source region of the summer high-level cyclone differs slightly from that of in midsummer, and is situated in the higher latitudes. Most of the summer high-level cyclones follow irregular paths and. sometimes the cyclone moves westwards. According to its path the summer high-level cyclone may be classified into following types:. Type I: moving westwards from South Japan to SouthiChina Type Ii: moving westwards over the ocean far south of Japan Type II: moving northeastwards from the south to Japan Type IIi: moving northwards off the west coast of Japan Type II/i: moving northwards off the east coast of Japan Type III: moving westwards over Japan for some period (path describes a loop) Type IV: appearing near or over Japan after the northeastward passage of a typhoon off the east coast of Japan Type V: moving eastwards along or off the Pacific coast of Japan Type VI: migrating in an extended trough off the east coast of Japan Types IrtvlV appear normally in midsummer and Typ.es VrtvVI mainly in the Bai-u season. The Type IV belongs to the one developing when a typhoon contacts with a high-level cyclone coming from the north. Generally speaking, in summer, the subtropical anticyclone belt lords over Japan, consequently in the upper troposphere prevai! light winds, or in other words, low zonal index circulation is shown. The summer high-level cyclone movement in these circumstances is likely to be steered irregularly by anticyclones surrounding it. Herewith typhoons steer it, because in the upper troposphere an intense typhoon acts a role similar to that of a subtropical anticyclone as being shown from its structure [2] [4]. According to rough estimations, the steering level is slight!y higher than in the case of the high-level cyclone in other seasons, namely 150-mbrtvlOO-mb levels. For example, the summer high-level cyclones developed in August, 1952 and 1955, describing loops over Japan were steered by surrounding anticyclones successively. Similarly, the summer high-level cyclone originated off the south coast of Japan in late July, 1959, moved westwards steered by an anticyclone on its north side. In the case of the cyclone developed in lateJune, 1959, its. southward movement was not expected referring to the pressure pattern below the 200-mb surface; a day before, however, the pattern at the 100-mb surface had already shown the tendency to its southward motion..

(8) T. Masui. 28. g3. Thermal Structure of Summer High-Level Cyclones The central pressure of the ordinary summer high-level cyclone is not very low; accordingly, contour lines on the charts do not concentrate themselves closely. The concentration is found most prominent slightly below the elevation of the jet stream. Closed contour lines on the charts appear from the 500-mb level up to the 150-mb, sometimes to the 100-mb level and towards the lower layers disappear gradually, except for the cases of Types V and VI. As for the latter two Types, we may see different patterns in lower layers, that is, the situation of the low pressure centre is displaced far from that of in higher levels and in many cases these patterns are not. '. e. directly related to those of in upper layers.. The characteristics of the summer high-level cyclone are indicated most evidently in its thermal structure. The cyclone following an abnormal path in the Bai-u season (Type V) shows a structure similar to that of the highlevel cyclone developing in.the cold season, with the exception that the prominent level is higher, namely at the 500-mbrv200-mb, and the whole feature 80. -.. mb. 100. -C5 -70 -C5 -65. ss. 150. -Ge. -60. -55. -55. -se. -50. --5. -45. ,3o. co. -4o. Ge,. Ns. -40. +t 11. 3e. -35. li lx. s1. -35 -30. t30. -25 :, il''''. -25. 250 '. 300. 350. '. ttlt. 1,s. -ze -t5. 't. '. -15. 400. -1O. b. -10. '. s. i-. -5. t''--s'L. tt. ,',,. t''lt. i.L. t. ss. LsN. 3e--GO-'. -5. 1. s. L. Oee-5so--. 200. 500. o. vLt. lt. 600. 1'. 5. 1. 700･ 800 .:o --850 15so900 20 .. --- --.-1000 1,O5O Aki XVinj YonFukKagNazKad lo. ii ',. ttt. s-s. ttlLbl. .-. 10. 1xs-. .. 1. i. -. tt. s.. o. ss. -.-t. Fig. 1. Vertical cross section along the line from Akita to Kadena at 1200 GMT, June 24, 1958. Thin lines: isotherms (in OC), thin dashed lines: humidity lines (in %), thick lines: main discontinuity lines.. ,.

(9) Summer High-Level Cyclones in the Region of Japan 29 ' is not so significant as in the case of the cQld season (Fig. 1). The cyclone. of this Type has normally a warmer portion above the centre and a colder one below it;this colder portion, however,' does not reach the surface, and generally, a weak frontal zone exists in lower layers. This zone is formed,. perhaps, by the descending cold current before the cyclone development. The vertical cross section of the high-level cyclone in the cold season is given in Fig. 2, for the convenience of comparison. 55. -5e1. -GO. -. ' 65-7o. so Mb -7o. 100. -65. 150. -GO. -S5,. 20o. "50. 250. -50. -+5. -o30O -35 -30 u25. 350 400. -50 ,-. -20' exO. -s. -15. N. ,-. -o. go. h11. -5. -30. .:. .s. t. 4t. jr. i. Ns. Ns,. -20. .-"-. s::. '. ss. 'so. N. 1. +. -i5-. 3o. '. '. -25. 500 66o. -10. S-''trFN. -35. --. tgz.. -s. ssO. oSo. 5. 700 800 850 900 IUOO 1O50 10. so. ' h. o5. 10. Whkgap Aki waj YonFukKag Kad. Fig. 2. Vertical cross section alongtheline from Wakkanai to Kadena at OOOO GMT, March 22, 1957. Thin lines:isotherms (in OC),thin dashed' lines: humidity lines (in %), thick lines: main discontinuity lines.. 'In contrast to this cyclone, the summer high-level cyclone developing in. midsummer appears main!y in the upper troposphere. As an example, we observe the vertical cross section of the cyclone at OOOO GMT. August 4, 1957 (Fig. 3). Though the origin of this cyclone was not clear because of developing over the ocean, this cyclone, at this time, was thought to be in its fully deve!oped or even decayed stage. As shown in Fig. 3, on.the] .$outhside. of the lower temperature portion below the centre, humidities are higher than those in the surrounding area, indicating the existence of upward current in. that portion. The higher temperature portion above the centre shows the existence of' descending currents, too. But, unlike the normal high-level. cyclone in the cold season, the high humidity porEion does not reach lower layers and is not saturated. The tropopause height in the central portion is.

(10) T. Masui. 30･. .,60. nib. -6. 80. e. 100 -GO. ,70. ･-7o. 150. --GO. -60 -55. -55. -50. -50. -a･s. --5. --o -35. ss. -30 -25. -+o. '. ,. S,O. lh. -15. t-. -le. '. 1. -20. ,1,tb. tli. -30t. '. , ,. ,. ll pl 11 Jl. ,tl. 1"L 11. tt tt. ,. ls. t-. -5. -25 ,. IL, s-. L. ,1. -30. gego3o-t-. L. -20. .Ni;s. -15 K--. .30. -o. '. '. -s. it. 5o. lt ,. ao. hlOj.Il sx# sNi. h". -ill '. ---. s-. -s. 350. 400. 500 600 700. ss. 15. NS NS t. 15 '. s`. Sap. 300. -10. ''. eos. l,N. 250. -35. GO. goC.O. 200. o. Ls. so. 25. 800 850 900 10oO 1050. AkiSenTatHachTor lwo. Fig. 3. Vertical cross section along approximately 14ooE at oooo GMT, August 4, 1957. 0ther legend the same as Fig. 1.. relatively lower than that of the sgrrounding area in the same way as the. caseoftheordinarycyclone. '' .t t.,i. As shown in the vertical cross section,, the thermal structure of the summer high-level cyclbne is far different from .that of the typhoon which has a higher temperature portion in the central part throughout all levels. The existence of another higher level tropopause above the cyclone is a significant feature of this summer high-level cyclones for this tropopause is not clear in ･the cold season.. g4. Development of Summer High-Level Cyclones As for the development of ordinary surface cyclones, it is said from the point of view of the synoptic meterology [12], that the cyclone development at sea level occurs when and where an area of appreciable vorticity advection. in the upper troposphere becomes superimposed upon,a slowly moving or quasi-stationary front at Sea level. Here the lower laYer acts an essential r61e in the development. But in the case of the summer high-level cyclone, we find no considerable relations between the upper and lower layers as described before; we may, therefore, consider mainly the development of the. '. `.

(11) Summer Hlgh-Level Cyclones in the Region of Japan 31 upper trough. The common feature of the pattern on the upper air charts, which can be seen in the case of trough development, is the presence of a lower-temperature and a lower-humidity area on the west side of the trough, showing an inflow of cold air coming from the north. This cold air shows a relatively strong descending motfon, as discerned by many investigators [911], and the order of magnitude of the vertical velocity is estimated to be about 10-2 m/sec.. With regard to the process of development towards the meandering westerly current, J. Bjerknes [1] has discussed it in detail, applying the equation of the vorticity change of the individual travelling particles:. ' d4-ap dt -aa ax ay6ip ny ay Oa a.-(C+2S2sin･￠) divHv. '. t ttttt. -(22C.OS￠)v+6aWy(22cos￠+aaU..)-aaW.g:･ (1) ' The first term to the right represents, in component form, the effect of isobaric-isosteric sol.enoids on the change of vertical vorticity. This term is. always found to be insignificant compared with the following ones. The divergence term shows how horizontal expansion (divergence) creates negative (anticyclonic) vorticity, and horizontal contraction creates positive (cyclonic). vorticity. The next term shows the effect on relative vorticity 4 of the dis-. placement of the air in meridional direction. The two last terms on the right hand side describe the influence of the vertical motion in changing the. vorticity about the vertical. ,. '. When 4 is positive (cyclonic), C.==4+22 sin￠ is large. On the other hand, when C acquires large negative (anticyclonic) values, the absolute vorticity 4a. may go down to zero or even become negative. This happens almost exclusively in the upper troposphere and lower stratosphere. where the wind velocities are very strong. On wave crests where 4. reaches values close to zero, the vorticity change is only feebly infiuenced by the horizontal divergence,. and obeys mainly the term of meridional advection in (1). This would be equivalent to a motion under the approximately constant absolute･ vorticity. 4.f:O or C=s-29sin￠. If a wave crest in the upper atmosphere has developed to that extreme stage, the parti' cles overtaking the crest would maintain their anticyclonic vorticity for a long period thereafter. Fig. 4 illustrates the process of successive (ltN/3) degerieration-of sinusoidal wavepattern caused by excessive anticyclonic vorticity bn wave crest. The final development does not always consist in the cutting-off'of a closed cold low. at the southern end of the cold trough. ' A vertical motion effect of the right sign to explain the growths of -C. beyond 22sin￠ would be found north of the maximum of upward velocity in the cyclone (Ow/ay<O, Oza/6z>O). The result in terms of a large anticyclonic vorticity, and occasionally a negative absolute vorticitY, can then.

(12) T. Masui. 32. ". t. 2 .. ' ,Q ' Fig. 4. Successive degeneration of sinusoidal wave-pattern caused by excessive anticyclonic vorticity on wave crest (after J. Bjerknes).. be expected to accrue on the upper wave crest to the east of the cyclone. Analyses of observational data do show areas of negative absolute vorticity on pronounced upper wave crests and/or south of pronounced "jet stream". To estimate the order of magnitude of various terms in (1), we shall begin with the computation of the vertical velocity of the descending cold air. In general, vertical velocities are computed by using the equation including the term in vapour pressure; in the upper troposphere, however, the. vapour pressure is low, so we may compute the descending velocity by means of using potential temperatures [10]. Vertical velocities are computed for each 12- or 24-hr. period, by taking the difference in height between the beginning and the end of each 12- or 24-hr. trajectory. We estimate the trajectory,,of the air particle by the wind direction and velocity of each level,' referring to humidities and potential temperature. The individual change in potential temperature' of an air parcel is given by. de. OO. oe. oo. dt=ot +"s os +W og, (2). where s denotes the direction horizontally along a streamiine. If dO/dt=O,. .. ' gf ][ g.e ]-i, .(,) .==[- g,o -,, ". . w=(gf),+v,(6og,),, , (4) where z is the height of the isentropic surface, and the subscript 0 denotes that the motion is isentropic. The last equation shows that the vertical motion of an individual air parcel is comprised of the 'local vertical movement of the isentrepic surface, plus the vertical component of the quasihorizontal wind due to the slope of the isentropic surface in the direction of a streamline. The method used here is simply a graphic way of integrating these two terms over the time period covered by the trajectory. In the cold season the estimation of trajectory is diflicult in the Far East because of intense westerlies. In summer, on the contrary, wind velocities.

(13) Summer High-Level Cyclones in the Region of Japan 33 are not high; it may be thought, therefore, that the estimation is･not difficult. But, when wind velocities are too low, the estimation is rather difficult on acount of the spacity of observational stations. Some errors in estimation cannot be helped in both seasons. J. Bjerknes discussed the first development of the trough. In the region of Japan, particularly in summer, the first development of a trough is rare;. we must, therefore, deal with the secondary development of the trough by successive cold outbreaks, but it may be treated theoretically in the- same. way. In summer the wind speed in the upper troposphere is estimated at about 20m/sec and the vertical wind shear at about 3m/sec per 1000m. Next we must compute the order of magnitude of the terms including the vertical velocity, and mainly considering the secondary development of a Ow 6u 6w Ov 6w av. trough, we may investigate the term -ox og inStead Of oy oz ox og'. The order of magnitude of the vertical velocity is estimated at about 10d2. Oab . m/sec as known from the following examples. The value of the term ax is positive in the light of the analysis carried out over North America [7], but in the Far East it is relatively difliicult to determine this value, particularly. in summer; fortunately, however, we could estimate this value in the case of a middle layer cyclone in autumn, namely 10d7sec'i [6].. Using the above values the order of magnitude of the terms in (1) is. determinedasfolldws: ･ 32 cos￠ Va 'vlO-`' 6 ?Olo, =-gr "10-9 seC'2. '. OwrvlOH`･10-7=10-'i secri2 22 cos ￠ Ox. Ow rvlOH7 av ･ 3 ･ 10-3=3 ･ 10-iO sec-2. Ox ag When the descending motion of cold air is relatively stf6' ng and the vertical. wind shear large, we can see trom the above values that the order of magni-. tude of the term OoWx gZ is comparable with that of the term 22cos￠ Z. According to this fact, we may expect a farther development of a trough and frequently into a closed cold low. In summer, however, the wind in the upper levels is sometimes very low and consequently the vertical wind shear. 6w Ov. is not large. The contribution of the term - ax az tOthevorticitychange,. therefore, is subjected mainly to the vertical wind shear rather than the horizontal gradient of vertical velocity, on occasion.. Now we consider the existing level of the cyclone. In summer, the meridional temperature difference is small, and extremely cold air masses do not appear in the upper troposphere over the higher latitudes, because the region of Japan is involved in the subtropical ariticyclone belt. In the lower layer.

(14) 34 T. Masui. the temperature rises, too, but the stability becomes high and the ascent curve on the adiabatic chart shows sometimes inversions in lower layer. Consequently, a dry adiabat drawn passing the point indicating a cold air in the upper troposphere over the station in the higher latitudes crosses frequently the ascent curve of the station in the lower latitudes above the lower layer. If a cold air mass in the upper troposphere is brought down adiabatically, it could not penetrate into .the lower layer. In summer, especially when an intense typhoon visits the iregion of Japan, the cold air coming from the north is brought into the rear side of it, with the result that the high sta-. e. h. bility in the lower layer is re-established. In the cold season, on the contrary,. the meridional temperature difference is large and the stability in lower layers is not high. If an extremely cold air appears in the upper troposphere over the northern station, the dry adiabat drawn passing the point indicating this cold air does not cross the ascent curve of the southern station. For these reasons we can understand that summer high-level cyclones exist only in the upper layer and in the cold season intense cyclones at sea level.. The high-level cyclone moving eastwards off the south coast of Japan during the period June 26rvJuly 1, 1959, was investigated as an example of development. On the 200-mb chart, the trough in which this high-level cyclone. was formed had moved southwards slowly from Central Manchuria since June 18 and on June 24 it extended from the Japan Sea to the South China Sea. At 1200 GMT, June 25 a high-level cyclone was seen over the South Okhotsk Sea and the trough extended from this cyclone along the northwest coast of Japan to the East China Sea. At this time, on the 200-mb chart, a wide area with cold air showing the temperature below -590C spread over the northwest side of the trough and the lowest temperature part.having the temperature below -610C was found over the region extending from Korea to the Southwest Japan Sea, but the source region of.the latter was not known for lack of data. On June 26, when the trough crossed Japan, it was divided into two parts; one branch extended towards Korea and the other towards far west along the south coast of Japan at 1200 GMT. And the area over West Japan was covered by a weak ridge extending from the west, although the temperature in this area was low. But in the middle layer (400-mbrv 500-mb) only the southern .branch existed and closed contours were seen faintly. The cold air which contributed to the development of the southern. branch came from Northeast Japan and the maximum vertical velocity of descending current at about -7.4cm/sec was computed using a graph as shown in Fig. 5 (height difference 3200gpm per 12 hrs between Sapporo and Tateno). But, probably, this cold air did not reach the 700-mb level because of the high stability in lower layers.. At OOOO GMT, June 27 the northern branch of the trough which 6ad ex-. isted in the uPper layer on the previous day disappeared and the southern branch of the trough developed still more and was outoffed by the ridge. 9. o.

(15) 35. Surnmer High-Level Cyclones in the Region of Japan. se. '. 1oe. :,. 11,'. 150. .//.. -6. 200. o"x. Nx. xNNx. 250. xNxxN x. NN' eN. 300. ts. -4O. 350. NNx. 400. NNxNNNN .. -20. eN. 500 NNxxN. 600 N･. xNl. 700 NoeNxN. 800. xxNx '. '. i200 lNN. 9Ue 1000. Fig. 5. Ascent curves for Sapporo, OOOO GMT, June 26, 1959 (solid line) an d 'for Tateno, 1200 GMT, June 26 (dashed line). Arrows represent the same d e･sc,ending air parcel (0==325eK) . Slanting th in solid lines: dry adiabats..

(16) T. Masui. 36. ` '. 80. i' 't. !/. w. '. ll'tl. 1OO. ,. ･1. ,N. 1. l. t.. ･N''Nl'Nl. ･N. 150. ,. -NNNA NS. -N .-. N.. NN.. '. N･ ts. 200. N600N･.. -. N. x. xN N. t'.'' s･.･'･.'.,'. N.N.N. 250. N.. N. N.N. N. .. Nx'N. N･ NN N' Nx. 300. -4O d'. 'x. 350. Nts.N. .. Y. 'N. 400. NNx'X. .oX.N. -2O NN.. NN. 500. .. l. xN. N`. .. N.NX .. ･N. N'. 6eo. N"･etN.. o". <. 7CO. N. ll.. IN.. XN.. NN. soe. Nl'1N'NNsl. 900. NN x 2. .. N.. 1OOO. oe. Fig. 6. Ascent curVes for Fukuoka, 1200 GMT, June 26, 1959 (solid line), for. Kagoshima, 1200 GMT, June 27 (dashed line) and for Naze, 1200 GMT, June 28 (dash and dotted line). Arrows represent the same descending air parcel (0= 3370K). Slanting thin solid lines: dry adiabats.. e.

(17) Summer High-Level Cyclones in the Region of Japan 37 moving towards east. The development of the high-level cyclone in the upper. layer was very rapid, but this sudden change of the pattern in the upper troposphere could not be explained only by the descending motion of the cold air coming from North Japan. The cold air which was closely connected with the development of the high-level cyclone, particularly with that in the upper troposphere,' was the one advecting in the upper layer from Korea towards North Kyushu on June 26. The descending motion of this cold air was caused. probably by the formerly descended air from the northeast. As shown in Fig. 5 and 6, the descended air, in this case, could not reach the lower layer, as well. Hence the high-level cyclone existing only in the upper layer. While. the cyclone was developing in the upper troposphere, on June 26rtv27, the wind velocity in the cyclone area was very low, and consequently the trajectory was relatively diflicult to describe. The vertical velocity of the descend-. ing current in the upper layer was estimated at about -1.8cm/sec. The vertical wind shear was small as illustrated in Tab!e III, and the order of. Ow az Ovwas nearly 10Hi'･ From these values, theremagnitude of the term ox fore, the intense development of the cyclone was not expected.. 'Ow ov. Table III. Wind shear and valUeS Of ox az ･ Fukuoka; l200 GMT, June 26, 1959. .Levels. o Winddir. vel. m/sec. 500-mb 5817 gpm. 400-mb 7508 gpm 356. 99. 2. 1. 350-mb 8478 gpm. 300-mb 9561 gpm. 200-mb 325. 11. 33. 8. 4. 3. 6. Ov (Sec-1) 6z. Ow Ov (sec-2) Ox Oz. 250-mb. 10791 gpm 12221 gpm. 4. 2.8 . 10-3. (Ox =100 km;. aw =1.8. Ox. . 10-s). 5.0 . 10-!i. Kagoshima; OOOO GMT, June 27, 1959. Levels. o Winddir. vel. m/sec. 500-mb 5840 gpm. 400-mb 7527 gpm. 74 2.. 59 6. 350-mb 8501 gpm. gpm. 250-mb. 200-mb. 10822 gpm 12264 gpm. 44. 70. 25. 4. 8. 9. 17 12. 2.9 . 10-3. Ov (Sec-i) Ox. Ow 6v (Sec'-2) Ox eg. 300-mb. 9586 '. 2.3 . 10-3 (ax= loo km;. Ow=1.8 . 10-s) Ox. 4.1 . 10-n. The front in the lower layer was brought out probably by the precedingly descended cold air which had no connection with this high-level cyclone.. Another:example which we inquired into was the summer high-level.

(18) 38. T. Masui. 80. `. 1OO. (. ii. '. NNOON. 150. 'ssNNxNN. 200 N'''xN. 250 x. xN. 300. NNNxx .. ---. 40. 350. NNx. 400 xxNNt. -20. o. 500. xNNN. p. '. Nx. 600. Nx. o. Ns. Oe. 700. NNN. 800. .. '>. N lNOON. 2. 9oo 1000. Fig. 7a. Ascent curves for Khabarovsk, 1200 GMT, July 26, 1959 (solid line) and for Wajima, OOOO GMT, July 27 (dashed line). Arrows represented the same. deScending air parcel (0==331clK). Slanting thin solid lines: dry adiabats..

(19) Summer High-Level. Cyclones in the Region of Japan. 39. 80 It -l.st. -. IN. 100. -. '.Nt'Nl. 't N･1NA tl'N.NN. NN.NN.NN. 'NN. x.. 150 x x-NN-NX'. - 60"N,. N N -. N.,N .N.. .. 200. xN. 'xN 'NN. " .. 250. Nv. xv. NX.. -40 oxN. tuN. 300. x' '. N. Y･NN. .. '350. .N. N. x.N .N. -2O eN.. 400 N'N･. N-ss. .". seo. N. NX. 'xN. .N N.. 600. ..NN.. OxN. .x. N.. 700. -N. N-XN NN -. NS .. 800. s.'. 900. ts.x tu. .N 2. .C. 1OOO. Fig. 7b. Ascent curves for Wajima, 1200 GMT, July 27, 1959 (solid line), for Shionomisaki, OOOO GMT, July 28 (dashed line) and for Kagoshima, 1200 GMT, July 28 (dash and dotted line). Arrows represent the same descending air parcel (0=3450K). Slanting thin solid lines: dry adiabats..

(20) 40 T. Masui. cyclone moving westwards from July 26 to August 5. The motion of the cold air taking part in the cyclone development is shown in Fig. 7a and 7b. The first deepening of the trough began in the levels 400-mbrv300-mb in the northwesterly current over Central Japan. The cold air which had existed at about 300-mb level over the area near Khabarovsk on 1200 GMT, July 26, descended, as a whole, to the level about 800-mb over the east half of Japan. e. .. and the most prominent downward motion (w =-4.0cm/sec) was seen above the 500-mb level. Following this situation, on July 27, the cold air near the. 200-mb leVel over Wajima which had come from Siberia, accompanied by a weak trough, flow-ed gradually down to the south coast of Japan, at the level slightly above the 500-mb surface over Kagoshima. But this cold air did not go through the 500-mb surface, so that the cyclone development was restricted in the upper troposphere. As we see, the development of this high-level. cyclone occurred in two successive steps. In the second step the cold air could not reach the lower,layer by the existence of the cold air which had descended formerly. The vertical velocity was computed using Fig. 7b, and we obtain the. value -1.7cm/sec in the upper troposphere between Wajima and Shionomisaki. We estimated the value of -0oll[l- assuming the range of the descend-. ing current (ax) as 300km. But the vertical wind shear over Wajima and Shionomisaki at 1200 GMT, July 27 was very small as shown in Table IV.. 6w Ov TablelV.Windshearandvaluesof Ox 0z . Wajima; 1200 GMT, July 27, 1959. Levels dirl o '. Wind veL m/sec. 500-mb 5873 gpm. 400-mb 7591 gpm. 350-mb 8581 gpm. 347. 347. 355. 6. 7. 7. 300-mb 9683 gpm 23. 40. 5. 7. 200-mb 355. 9. 1.6 . 10-3. (Ox=300km: g:'=5.6.10-8). e. 9.0 e 10-ii. Shionomisaki; OOOO GMT, July 28, 1959. Levels wind gLi: fuisec. 500-mb 5876 gpm. 400-mb 7588 gpm. 91 6. 94 16. 350-mb 8577 gpm 97 17. Ov (Sec-i) Og. 6w Ov (sec-3) Ox Oz. ". 6.4.19-3 2.3.10-3. Ov (Sec-i) ag. Ow Ov (sec-2) Ox Oz. 250-mb. 10935 gpm 12395 gpm. (Ox=300 km;. Ow= Ox. 5.6 . 10-s). 300-mb 9683 gpm 69. 250-mb. 200-mb. 349. 347 13. 10942 gpm 12436 gpm. 4 -,.4･atO-2). 9. (1.9 . 10-io).

(21) Summer High-Level Cyclones in the Region of Japan 4t From these values will be deduced the order of magnitude of the term. ow av 6x 0g. which is about 1.9 ･ 10-'O sec".. This cyclone developed still more by the successive downfiow of cold air, while it moved slowly westwards off the south coast of Japan. The maximum development occurred over the East China Sea on July 30, but since July 29, when the cyclone had moved into the East China Sea, the vertical motion and the temperature distribution in the cyclonic area became insignificant. The process of the last development, therefore, was not known and the process might be connected with the development of the easterly wave, because the development of the cyc!one took place in the easterly trough. Though it is not evident for lack of data, about the 29th, there might be an infiow of cold. air from the north into middle layers over the East China Sea. The movement of this cyclone was steered by easterlies to the south of the upper high. which grew gradually in the upper troposphere. ' The process of development of this cyclone resembles that of the easterly wave from the extended trough as indicated by G. C. Cressman [13].. g5. Examples of the Summer High-Level Cyclones ExAMpLE 1. Westward-moving high-level cyclone through the East China Sea, July 27rtvAugust 4, 1959. Type I. Figs. 8a-8f. The source region of this high-level cyclone was in the upper troposphere. over Japan and this cyclone moved developing westwards from the south coast of Japan through the East China Sea and dissipated at last over South China. The perfect westward movement of acyclone like this is a rare case and for the purpose of researching into the process of development of a highlevel cyclone it is convenient to us that the source region was situated within the Japanese observation network. As for its development, the author dis-. cussed in S4, but he could not look into the whole process, because the maximum deepening took place over the sea and the cyclone decayed over South China where the observation data was relatively sparse. It seemed that the first development of the weak trough in which'this cyclone was formed, had taken place several days before, perhaps, in the northern latitudes. i The synoptic situation during the period of the development is as follows: , At 1200 GMT, July 27, a relatively weak trough began to deepen in north-. westerly current over Japan on the 200-mb chart. This trough progressed slowly southwards, developing into a deep trough extending far westwards. Meanwhile, a weak typhoon moved over the Pacific Ocean southeast of Torishima. This typhoon followed a very erratic path influenced by the anticyclone which was formed to the south of the trough afterwards. A highlevel cyclone developed south of Shionomisaki in this trough at 1200 GMT, July 28 and it moved westwards. The lower temperature area on the north side of the trough moved westwards, too. This cyclone progressed westwards.

(22) T. Masui. 42. h20'. ilO. ,,o. t+o. kJ. ' `'f" 5-Xxtrrev. L'it. --. b. ' ' ･xV'. +o. 4o t.. 6t2N ix ,.,6. .. p. 6i2 /f }pt-pts-i2. e tt. 29YO29-DO 6iLttiLo-.,o. 3e. 2s,;,6>'CP2-,,. f. 0. '. 29-OO. 3a ------oO030-I2. 3e-12. e. sgK. 29･t2. S 3-12. ×. .). 20. S 2-12. 20. o". Fig. 8a. Track of the high-level cyclone, July 27-August 4, 1959, on 200-mb charts. O symboles indicate 24- or 12-hour positions. Typhoon symboles indicate positions on surface weather maps. Plotted number groups indicate date and time (Data-GMT) of positions.. ne. lao. l2+OoX. e 1+e -N,LL )! '. i70. ･ lzQoe. s{i `50x, @,>.< w' Ss<v. ..-'Y<.NNtslll!z '--- 1. Ns t x-SNIS.. 1IxN '-50 4 :'::[email protected]･. w <l[I. H l. >. t2200 .. -e. QpmY <S----ss. g. ..--' @ff eeeti ua. s. '. H,/,,.' "----t' tu!. 120. -5o. ･ e.,･i･'". /.x4.. tt cJ 1/ x. /. :',ljli.'; i+e. }se. 12-OO. s- --t. B. fe. r 2e. z'. see. 3". o. xxxx(II. c. ×----. 20. 200-MB CHART 1zc. br. 110.==S. Fig. 8b.. tse le. )se. 200-mb chart for 1200 GMT, July 27, 1959. 7GMT-JuL,27,1959x.

(23) 43. Summer High-Level Cyclones in the Region of Japan lte. ,3o",, eN.tt. 12e. ..,ljotl,. 'JV, .,,{so. )70. IGO. ,･k/ (,. HW (r .. ･>. XN. >. y, <x IEil,. ttt. -50. SN ×. lttt. 120 } 1. . H7/,t. '. f'. lo s. ''. N NN. -tro- ----.. 12200. 7u] )-.-i..V. -t =--". t,<. til 1-'SO. tttt - -. o. ¥2o. CD. '3" lt". 12+eo. ;3o. 12+OO. ise. t=. S. ( ,/g. clLi'. --" / ' VJ/" ,A'Z. N. y. ,, Xl/:1 N"N. 2fi,L) /bfi. N"s...2 "-/. '. <7. g Eg, K, si., ,e,/}i{.,c･z･rk. 7'/ .-fss. } it--L-. 1). fi /tNx. v. -,tr .;S9g. lt. H. 10 -55 iv.. 200-MB CHART 1200 GMT-jUL28,IS59. i; 7. mo a. IBa. 200-mb chart for 1200 GMT, July 28, 1959.. Fig. 8c.. l20. LLe. ISO . lgO,. 1-O. tgw ,.sL/ i6e. Ae. L30× q. vi22ee. {g(1,)･N. ti40D. L. ilL{. /. H. tzo. f. '--･-+.--5. "g(--.." t --- -L.-"-50 p},ln<. iiiiiiiiii. /"-. A. Nl xv t. /. ti. g th -Rin. Ns. -q .. xE. sv. .U2. LI. N<O. -50 .. c-lll'. s.-..st-. k. o. s. N. -)of. ts". 30. tw. N. 12+oo..3O. vvib. 124. ll,-,"/i. lse. "e. -Es,i[. L. ss-J. y. <× '. y it. ttt. c. 1. 20. x. tt. t. t/. N. -55. xy' -tt. /t. Nx. 20. -200-MB CHART. i2o h. l30. Fig. 8d.. te. 200-mb chart for 1200 GMT, July 30,. 1200 GMT-JUL.30,IS5S. l50. 1959.. tce.

(24) T. Masui. 44. /. "o. -60----... -v--/-. FtA....t. m. ,. )'. ". g.y--'---------. J. ･-' '- (. --'. -6s. .. ae. 'Xr ---' ----" IGsoo. -70. ./e .. ln". '. I .r>nc･--t"-"---'. i'-. x -i/. ---- -60. v/. U x... 'd. J7a. lGgoo. -. 5C･-' s. ILx.. sto. N. s. H -GS---. LtO[ NvlSO s. /pg--lii-i-'. tse. ma. >. M'/ /-"Ntw' -×,J"::'ssN. v･.ViX. KN. 3.,F.wh ,". e. .,. l2ozl' 3o -7o. . -S9' -L.. ]. l' ,ki-{". IS'i '".. ..V/f7" S'. -- .4 iv.w.-"f i '/. tGD. 30. -t. ts". "h -tll/ ..x. i. lccoo. .. " .. -70. >. N. 'v'. ×. ?o. .75- L -･. 70. e' i --. -----. ･/. - 75 --. c. l80,`,MB,.C."."JuR'L.3o,igss. or" 1. t30. 100-mb chart for 1200 GMT, July 30, 1959.. Fig. 8e.. l30. IPO. LhO. '+. -CO, f. . --es-... '. SN. ･v'. i:r). ( 1.--Io. l. KFx-}2(.. (- ."･. J. teo. ...-'. )℃ 7 3o. A/ 7kn. IC800. re'v'S '". s. -N. 7. '. -. LC... v. =-itNY. c. <1. N. ePTilll. le. i50. t-lis". J30 CtSii. Ns. i L.. --tt"t-. t-. '. '. - -7o ---.... -- --70. ... '. -"pt. /-. 16COO ..- 75. 2o ,J g 'sN... yl'>.;s---.... ATS. 4. . -es. t!. /v /x, -1N"s.. dXrl - v4. tt. "". 40. N. '. ). ll. tx. -". -- -･ -- -･6O. n)" .. ... -.. II. ,i. K. 'N-,. '. ･x t.,S' Lt.,-1. c{. t.. X 7'(. .r. NNs-ss-. LL.. ! x. -70"+--. g. 20. ' '. ' '. '. '. c. 100-MB CHART 1200 GMT-JUL.31,ISS9. t2V wtFig. t30 to 8f. 100-mb.chart for 1200 GMT, July 31, 1959. ･/. t7e. IG600. ' ss X. i. I. '1. ltl. .. tae .T, J. sv. -t.. .70hNsx H. ;. /. ,O .A i5 I･. Ns' N l6800 .=･--l- b' m7o.,s "(. yfXL c.. -eS-'--'. c ]t. lso. iso. 14･O. t+o. Liee.

(25) Summer High-Level Cyclones in the Region of Japan 45 in 8 days or so following. At 1200 GMT, July 29 the cyclone was situated between Kagoshima and Naze and entered the East China Sea thereafter.. The cyclone reached its maximum depth ha!fway between Naze and Shanghai (at about 29.50N, 124.50E) on July 30 and everi at the 500-mb closed contour lines were indicated. We could not know how high the closed circulation was extending upwards, but at the 100-mb a wave trough in easterlies. was shown above the centre at the 200-mb. It seemed that the cyclone progressed westwards in this wave trough. At 1200 GMT, August 1 the cyclone. reached a position east-of'Shanghai, continuing westward-movement. On August 2 the cyclone was centred south of Shanghai at a distance of 200 km or so and after this time the exact position of it could not be found for lack. of aerological data. It is extremely probable, however, that the cyclone would have moved westwards with fi11ing. On July 29tNv30, a high-level cyclone was indicated on the 200-mb charts north of Iwojima, but we could not discern whether this cyclone vLTas the upper part of the typhoon or not. Throughout the whole period, in the area where the cyclone had passed, a large-scale anticyclone existed at the surface. and the cloudy weather predominated. Meanwhile, the centre of the subtropical anticyclone at the 200-mb was displaced from Korea to Central Japan by way of the Japan Sea, and another anticyclone over the China Continent was situated slightly more westward than normal. The velocity of this cyclone was comparatively small and it seems that the movement was steered by upper easterlies near the troposphere height. The wind speed in the upper troposphere was generally slow, not exceeding. 30m/sec over Japan. The typhoon described above was following a very erratic courSe.. ExAMpLE 2a. Northeastward-moving high-leve! cyclone from far south toward the east coast of Japan, July 30tvAugust 5, 1957. Type II. Figs. 9arv9c.･ This cyc!one was situated in a position convenient to us to study the thermal structure of the summer high-level cyclone as shown in g3. Though the time when this cyclone developed not clear, we can imagine that it. might have formed in the easterly wave trough. The development of the system on the 200-mb charts is as follows:. At 1200 GMT, July 30, a cyclone reached a point west of the Guam Island and then it moved northwards. On August 2 it was situated near 260N and 135.50E, and its closed pattern was seen at the 200-mb. On August 3, we could see its closed pattern on the 300-mb chart downwards and on the 150-mb upwards in the neighbourhood of a point 31.50N, 1380E. On August 4 it reached a posltion south of the Izu Peninsula and at 1200 GMT, it was situated near 360N, 1420E, moving northwards with fi11ing gradually. The cyclone moved along the west coast of Oou on August 5 and at 1200 GMT it.

(26) T. Masui. 46. M// o i3o. y t-e ,,,e,. i. - Lso. i. +o. XO 5-OO. yO+-12. .. ･). kcP 1O--oo 3-12 O. 3e. 30. a. f. O 2-12. /. O1-12. 2it. 20028-12 / .ifgs(-t2X. 031-12. 3esY to. 10. l+. Fig. 9a. Track of the high-level cyclone, July 30-August 5, 1957,. oni200-mb charts. Other legend the same as Fig. 8a.. t20 L.-.------./1,. t3" ---g :/, ,]kO( 1",'i)Z,, c `SO. no. oo. 1GO. ×. e. Q. /l/pa .di'v. L/c /r u./,v..,",Nil]9X-.-. I` eo. -5S '. x. t. ,. A". H 7'III]`' mao6. ()5 y>. k. io. jrr:i''I i CSI' ttiT=L:5t(Ei" -l',IIF'hx. X<> 12"ooh' N. x. ieoi ' W - Itl. N. NN NN s. e,. .･/ y. {) t2toe/ '39 Lv<o. 30. 15e. H -'`$SS,. J. H ,i'a P. Ns.$./ 0.J"' IGO. ×sL.. "v.. u. vt'-XJ' -fts. ,･K 'X -so.. 2o. 20 200-MB CHART 1200 GMT-AUG. 3,ISS7 ･. ISO. Fig. 9b.. 1-e. ts Je. 200-mb chart for 1200 GMT, August 3, 1957..

(27) Summer High-Level }!o. H. ,-5S. Ns'. t. < -sg.,. SXtN. iJ... vo "' 'V -. {gi,, f. -70 NNs ,N. ..lt 'H /i2o ISs x. A- n. ). li. cS. ". .. -F - -70. AD. f-'. l. C ;70. v':. 7'. IGO. 3D JiO. "l. . -/× `"""). MIG800 t･f. its, N. /. Jlv.fi7v// ,/A. xX ) L. ,7. ". -Cs. r:jl.(..;;'i-LJ`sv.-,rl. -rL-･-x' ....eg. s2. Y 16GOO. Yc/. ,,i. ,". 17S. Il$li. w. -i. <. .7sJSSOO N, tx. NL ti c,, / il. >'. s"1. Nk. tCo. L. :l. ....-----. -#f-i/ .f NV' .7'. 1. )'. <. 47. a. IN. -co. "7Sy c2. -ses-S3"W L. ipo. ,. Cyclone$ in the Region of Japan. c. N. NNt. ise H -t-70. tw. >. 20. 20. iOO-MB CHART 1200 GMT-AUG. 3 1957. n"L'i' i. ,. a3o. 1. Fig. 9c. 100-mb chart for 1200 GMT, August 3, 1957.. was lost in the high-level cyclone which moved from the west towards Hokkaido. 1. 0n the surface maps during August 3rv4, we could see a subtropical anticyclone extending westwards from the North Pacific Ocean and a front in the Sea of Japan. After August 5 the front removed away. The weather, however, was not bad during the cyclone movement. Meanwhile, at the 100mb a trough in the westerlies had moved from the Yellow Sea towards the east and reached Hokkaido on August 5, showing its closed pattern just west of Sapporo. It continued to move eastwards thereafter. In the easterly region a trough south of Tango was displaced westwards, reaching a point east of Kadena on August 4, and stagnated thereafter. On 1200 GMT, August 5, this trough retreated to the neighbourhood of Iwijima, reaching a position west of the Marcus Island the next day. Thus the high over South China moved eastwards and probably the high-level cyclone wa$ steered by this eastward-moving trough.. ExAMpLE 2b. Northeastward-moving high-level cyclone, August 5rv13,. 1958. TypelL Figs.10a-10d. .. ' way as that This high-level cyclone moved northeastwards in the same of Example 2a, but its initial development differed widely. In this ease the cyclone was formed, to the east of the Philippines, in the southern branch of the trough which had been disp!aced frotn the west. The northern hranch of.

(28) T.,Masui. 48. so lse :tox?Ol-.tio / lr-X., ,NO 12-1×2 O 13-12,1). x. 50., t .. C-120. ,.1>L. ,.,,. L f J">y. o8-12 4o t s"s -i' o(l-.,tt-#,cl,,x'3i,oo. v/. O11-Oo. ･-. 3o. 30 e f /OIO12. /i. 09-l2. ipt.:lil)f l/9 17T2'O 2e. 7-12 o le ,. l. so. Fig. 10a. Track of the high-level cyclone, August 5-13, 1958,. on 200-mb charts. Other legend the same as Fig. 8a.. if1 'LSO.. '7J. lfO. t+,o. / f. W. <. /. 12200 /. 1.,l""× <yt?. it/. ttt. Zx 7 6 c. ,. < "c.;S5I, IO. tl. 1. '. -. r. E. ke. Y (h.. N-. Bl. v. r. G. .V T2 fi----･;K.SS-z. geec. g<. ,. i)z. cr. s. l2o :;. 30. fio. y..,.--. U'. VUk.--. '. v. 30 ICa .. Ns. x. 'VN x. /. l. -55. c. l. :55. t12+eo ,. lse. 12S-oo N... tlt. /,. ,. ) ,S. YL.-ss.. ---. 170. /. y. x, g. tt;) -e ./. ',Se. 8. lse. ISO. H ' 2e. e. 200･MBCHART ' 12OO GMT-AUG.11,195g. ty, r2ollY eq ' 10. Fig. 10b.. l. so Tte. 200-mb chart for 1200 GMT, August 11, 1958..

(29) ) Summer High-Level Cyclones in the Region of Japan. tTO ylSOW. L3.0 / 1+O-cLo.A'X. l50 ess>llslgs. Y /. -S5 ... -gON. 16e･. t70. /. X lks , ss.-h-.-. ... N. -cs . ;-SO N.<'. 49. -. N･`'v --is-ige5(ISI U'<gN;. --'s W>}Nx. xx,( ･, L-,･;'. :ii s"/txs >i iit. t Nt. (tv):.rt'.A,7,i>x C t;-'. ' H J2o.) 3o XNIIIili. ""`-"'iivl "×, ;,. Z i3o-. .･. "･. '. '. -L. t. 70 3o le". .Ise. ' '. -ttttt. NN Ns. L -7o". × s.. 3!O. all]c ss-7o 16goo. ' t. sN. .lrt s4g>. -0 f ft. × --c;) . '. C 1660o. 1.. xu. 16e o. 100-MB CHAIU' 1200 GMT･AUG.10,1958 ,, !c, N. so. +o. tlo. 13" 1-O. .xsi ,,(t,"'Ylill 5(-,6 iso. t20. w. -55.. '. ,t.k. '×-<l/"'. -CO. IesoD. -6S..-!-- -. Q oO-. N, ";. , tt. l. fi. tl. ' '. '. 30 ,. .-- 7. .4-x. A. '. '. lt. '. v･. l30. ?.z(ll. .,,h-.r. -o. 1. -70 " t'AS:. -/. Xix s" 'y/ ,fL-:Kitk>. -.;t;} lttl. t?o...f!". J' V-･･v-v..v. Ktt･. t:. lSsNN. NSe. 3". 150,'. -t t. t e. -70. - .:1- .. -.. v. "'--"-'. /. ) IGGOo. ao. 20. r'. l3e. Fig. 10d.. IG600. 1OO-MB CHART 1200 GMT-AUG. 11,ISS8. c. 4 >'A. i2oL. .t7,O. -C5. !"M-- S' i...v.]. c. .I60 d. ,'1. 2 fo !S. -7or.. H. co. 100-mb chart for 1200 GMT, August 10, 1958.. Fig. 10c.. 1.0. LSO. 100-mb chart for 1200 GMT, August 11, 1958.. 'IGO.

(30) 50 'T.Masui ･ /･ '1 ･''.. '. -t. the trough which had moved through the Japan Sea stagnated on August 4 when a high-level cyclone coming from the north reached the .Kuriles. This high-level cyclone continued southward-moving and, on August 8, developed into an extended trough combing the high-level cyclone described in this example. The closed pattern of this high-level cyclone at the 200-mb might have been seen on August 5. The cyclone stagnated till August 7 off the. .. ". south coast of Japan in the large-scale trough described above, then progress-. ing northwestwards till August 9. The movement of the cyclone on the 200-mb charts is as follows:. At 1200 GMT, August 7, the cyclone was situated near 170N, 129.50E, moving slowly northeastwards. It reached a position at about 230N, 133.50E. on 1200 GMT, August 9 and then stagnating there. The cyclone to the north moved far eastwards the next day and this cyclone entered a new trough extending southward from Korea. After this time the cyclone progressed in this trough. At 1200 GMT, August 11, it was situated near 310N, 1360E, with deepening, and closed contour lines could be seen at the 150-mb. upwards and at the 500-mb downwards. The cyclone was centred to the north of Choshi, with decaying, at OOOO GMT, August 12 and progressed off the east coast of Oou thereafter. It became shallow gradually and might have been mirged with the high-level cyclone moving from the north towards South Sakhalin.. At the beginning of August, an anticyclone at the 200-mb was situated over China. This anticyclone crossed Korea and reached the South Japan Sea on August 3, but it moved back to the neighbourhood of the Shantung Peninsula on August 4, and then it moved southwards till August 7 when it reached Shanghai. During the latter period it was displaced northwards with the progressing of the cyclone described above.. , At the 100-mb, a trough passed over Japan during August lr-v3, then. .. stagnating off the east coast of Japan till August 5: but it moved back there-. after. On August 8 a new trough appeared over North China and during August 8rv12 the trough moved slowly eastwards. It is sure that the cyclone. was steered by the trough movement.. '. '. ' ExAMpLE 3a. High-level cyclone, August 23N27, 1955, moving westwards over Japan for a while. Type III. Figs. 11 and 12. On the 300-mb charts, a trough, penetrating far into the west-southwest, advanced southwards and entered the Japan Sea on August 21, when a high-. level cyclone was formed in this trough to the west of Hokkaido. Though the cyclone moved away towards the east of Hokkaido on August 22rv23, the deep trough still remained in the same region, extending from the Japan Sea. to Korea. The high-level cyclone which was dealt with in this Example, might be formed in this deep trough on August 22rv23. A surface front accompanied by this large trough move towards Pacific Coast of Japan during. v.

(31) Summer High-Level Cyciones in the Region of Japan. 51. Jue ii. 22 :,S2git15. 2 6-l scf' N>. ,/JL>. tt` oi ,p-g-Js. / 6--- 21.,,. .Iliill,lll,ii, e. i/ltNS75 t3o le. 20 --. Fig. IL Track of the high-level cyclone, August 23-27, 1955, on 300-mb charts. Other legend the same as Fig. 8a.. the same period. The process of development of the cyclone at the 300-mb is as follows:. 'At 1500 GMT, August 22, the cyclone appeared to the west of Waji'ma with the southward movement of the trough and it stagnated over Central Japan near Nagoya on August 23rv24. Meanwhile, a westerly ridge moved from the northwest towards North Japan and it became an upper anticyclone thereafter. The cyclone moved westwards influenced by this anticyclone; accordingly, it was outoffed perfectly. The cyclone moved towards Osake on. August 25 and at 0300 GMT, August 26, it reached the coast of San-in, entering the Japan Sea thereafter. On the same day it would probably have reached Korea; meanwhile, a typhoon progressed from the East China Sea towards Korea. It seems that the cyclone might have been displaced northeastwards in advance of the typhoon. It is thought that the development of this cyclone is caused by a cold air descending from the north into the trough. As illustrated in Fig. 12, by means of comparing potential temperature on the estimated trajectory, we could know that a cold air existed over the neighbouring area of Khabarovsk. at 1400 GMT, August 20, descended over Korea near the station No. 208 (37.50N, 1270E) on August 21. The process was still lasting on August 22.. The cold air which had descended from Hokkaido after August 23 had no relation to this cyclone, and, perhaps, the cold air, after descending from the north, circulated in this cyclone. This situation could be known by the dis-. tribution of humudities which took place in the middle troposphere in the. e.

(32) T. Masui. 52. `. 80. ij. 1OO. 'N'･N.tNts'NS.ss'N'N.,NS.Ns.N.N.. Ns. 150. -!!!-6oe NNsN･NN.NN･sN･Nx. 200. x-N""NN.NN.NNlXN. 250. N+Nx･xN. 'x.N.NN. lo. -40. 300. NX.. XN N'. NN.XN.N. xNx. 350 .N.N. 400. ,. .N.NN･NN,x.x.NL xxo. --. 2O. 500. ･x. NX. .. xx. 600. -N.xx.XN.NN-. ". 700. :eS"L.ONNN. soe. .. Nx. N. .N.N'. 900. .. .x.x.. 200C'. '. 1000 '. Fig. 12. Ascent curves for Khabarovsk, 1400 GMT, August 20, 1955 (solid line), for the station No. 208, 1500 GMT, August 21 (dashed line) and for Itazuke, 0300. GMT, August 22 (dash and dotted line). Arrows represent the same descending air parcel (0=3280K). Slanting thin solid lines: dry adiabats.. '. ,.

(33) Summer High-Level Cyclones in the Region of Japan 53 '. central region of the cyclone, that is, in the lower temperature portion humidities were low, too.. The development and path of this cyclone are reconsidered here, although. these were already discussed by Dr. S. Orihata [8]. . , ExAMpLE 3b. High-level cyclone, August 12rtv17, 1952, moving westwards over Japan for a while. TYPe III. Fi'g. 13. This cyclone belongs to the same type as the one described in Example 3a. On the 300-mb charts, it was formed at 1500 GMT, August 11 near Akita in a trough extending southwestwards from the high-level cyclone which had moved from the north towards Hokkaido. The cyclone went southwards at first till 1500 GMT, August 12 and the4 stagnating over the Central Kanto District. for some time. On August 14, it moved westwards and gradually curved towards the northwest by way of Kanazawa. At 1500 GMT, August 14, the cyclone reached its maxirnum depth and its closed pattern could be seen at the 500-mb downWards. It recurved northeastwards from its position, west. of Wajima, on 0300 GMT, August 15 and crossed the northern area of Oou early the next day. The cyclone was lost thereafter; meanwhile, a typhoon. hadreachedtheEastChinaSea. . ' . l30. t2a. t. lso. o. s. 2 IC-03. (y(91. is. / 4.9. 1+.o>a. . g {{rw. l);'z. o>o W5. IG -15. 40. -hSL. is-o3d. -"V･-X. /. 12-. l3-03 12-1sl-rS. v 13-15. tt. lt51G'15 , ･,''. x'. 3e. ,3o. S16-03 < , S15.-1,s ,... N .J is-o35. x IC -15 S. N. 5 i+-g3 ･i. sJUE5s.h- ,,. 20. 1/ -, 1-O. lso. Fig. 13. Track of the high-level cyclone, August 12-17, 1952,. on 300-mb charts.. Other legend the same as Fig. 8a.

(34) 54 'T. Masui '. On August 9, before the cyclone began to develop, a weak typhoon had moved eastwards from the east coast of Japan. It seems that the elongated trough, in which the high-level cyclone developed, was formed by the combining process of a high-level cyclone coming from the north with a weak typhoon. Similar examples to this were found during the periods August. .. s. 7rvlO and October 18nv20, 1953.. At the surface was a front stretching along the Pacific Coast of Japan and caused a slight rainfall during this period. Meanwhile, an anticyclone. moved from the west towards Hokkaido on August 1 and another extended. fromtheNorthPacificOceantowardsthesoutheastcoastofJapan. , ExAMpLE 4a. Eastward-moving high-level cyclone off the south coast of F9.gg.nla4ftaenrdSIOsU.thWard-MOving, June 21rv27, lgss. Type v.. The development of this cyclone on the 200-mb chart is as follows:. At 1200 GMT, June 20, in Central Manchuria a trough extending from the east went on moving southwards and a high-level this trough over West Manchuria the next day. The southwards and on June 24 it entered the East China east thereafter. It moved northeastwards for some reached a point off the coast of Shikoku. After that. cyclone developed in cyclone moved slowly Sea, curving towards hours and on June 25 it went eastwards and. 9hnerJeUanfete2r7. arriVed at a POint southeast off the coast of choshi, being lost. ' r')"" × l2o tgo wolsa. so. ,1 '. ,1. tL. l 021-12. -/" ,,1. N. o22-12. .. .wt.n.s'j li. Xo 23 -12. 26.e(tWLI`[ o7"9.9･7. i'h16cp t'`. 1. di. ×. 3o. 26-12. 5-12. 25-OO. 3b. 2`-'2. ,. f. /o･ t' ". '2o. 2e l40 Fig. 14. Track of the high-level cyclone, June 21-27, 1958, on 200-mb charts. Other legend the same as Fig. 8a. TIO. .. k". 2. -o. /". de.

(35) Summer High-Level. Cyclones in the Region of Japan. 55. x. 80. ttlll. v 1OO ssNNNsNNN. L. 150. NNNO".NlxN. 200. NNxxNx. 250. .N. NNN4oeNx .. '300. xxx' x. 350. r. xxN 400. x2o"xNNxx. 500.. xN x N. 600. xNxx Oe. .700. xNN. 800. NNx. 900 NN. 20'. 1OOO. C. Fig. 15. Ascent curves for Vladivostok, 1200 GMT, June 24, 1958 (sol'id. line). and for Fukuoka, 1200 GMT, June 25 (dashed line). Arrows represent the same descending air parcel (0==3440K). Slanting thin solid lines: dry adiabats.. tt '. ' '. '. t tttt tttt tt. '.

(36) 56 T. Masui. At the 500-mb a cyclone was situated over South Manchuria on June 21 and continued its southwards movement till 1200 GMT, June 23. Then'it. .. curved in the easterly direction from its position at about 300N, 1250E. It. arrived at some 300N, 1300E on 1200 GMT, June 24 and it changed into a weak trough extending from the northeast after reaching a point to the. '. south of Shikoku on June 26. Closed contour lines at the 100-mb could be seen only at 1200 GMT, June 24, to the east of Kagoshima and thereafter the cyclone at the 200-mb moved in the area under the trough at the 100-mb. During the latter half of the period, the anticyclone at the 100-mb which had been persistent over North China, began to extend rapidly eastwards. On the surface maps a cyclone, accompanied by a front, progressed from the southwest of Formosa toward a area south of Kyushu during the period June 19rv24. At 1200 GMT, June 24, when it crossed the path of the highlevel cyclone, it reached its maximum depth (998-mb) with a heavy precipitation to the east of Kyushu. The surface cyclone and front decayed there-. after, untill they dissipated at Shikoku on June 26. 0n June 26N27 did cloudy weathers dominate over the south coast of Japan. The first development of this high-level cyclone was caused by the cold air descending from the upper layer over Siberia. It is evident from Fig. 15 that, after the eastward movement of the cyclone from the East China Sea, there was an inflow of cold air from the Northwest Japan Sea. After June 26 both the upper and,lower layers in the cyclone were separated from each other and during the latter period, the upper layer 'was isolated as a highlevel cyclone. This type of the highzlevel cyclone appeared frequently in the Bai-u season. Another example is what developed during May 31rtvJune 3, 1958 and before its development the synoptic situation of the upper troposphere was found very complex.. ExAMpLE 4b. Eastward-moving high-level cyclone off the south coast of Japan, June 26rvJuly 1, 1959. Type V. Figs. 16artv16d. The development of this cyclone is discussed in g4.. ExAMpLE 5a. High-level cyclone, July 6rvll, 1958. Type VI. Fig. 17.. On the 200-mb charts this cyclone was formed in a trough extending from the north-northeast on July 6, fol!owing a very erratic path thereafter. Its development at the 200-mb is as follows:. It was once formed near 370N, 1350E on 1200 GMT, July 6, but, afteryvards, was lost in a trough till July 8. The trough Ipoved eastwards crossing Japan and deepend rapidly by an accelerated southward current on July. 7, when the eastward movement of the upper anticyclone on the west side of the trough stopped. At 1200 GMT, July 8, its closed pattern appeared again in the trough. The cyclone then moved southwestwards and on July. 1. p. o.

(37) Summer High-Level Cyclones in the Region of Japan. t30. lf 'uo. YT. 57. h50 " .l fX)32 2C-12 /o 2E-OO. " q,}). ,lzx.t'-!2 2a2 >so. k. -o. ;3. / o ../ r .-. cb ex7";7':". K.. 26-12. ×o-611LO 2S-12 (. 5e. 27-12ONO.2S-I2. Q3on. f. 2. 2oJ2e k 11e. Fig. 16a.. on. l50. o. Track of the high-level cyclone, June 26-July 1, 1959, 200-mb charts. Other legend the same as Fig. 8a.. "e NY ,. 5ess ciillliili,55 'K,('"b e.. ieo s7o. Lw a. N. -"g. .g...･. ...-i.-. ,". A ,/ ,1' xNlNi2ooa. -55. '"'tfO..!.. ../'irt :' ,ili'. +o. "･/4..e, tt"ll- S. .,/ki, 'l,e.i,i,,i{･･i`o. tll-. w. :,. SNss-. -..-'. H. Nl ･,.J' t 12 -,'. VV' vilt l/. ... ..- 12j. tl tlt Ls."-N :!sp---------- --.s. Io. V･･ ,)' ･12A-OO. 'A f- ,?. .L!''L )'s'o". J55. f xe. l /. .1 t.t t-v-tt. '-. Jl y'. A. 20 .. H. 10 200-Sv!B CHART. 120O GMT-JUN.26,lS5S. i,･ r..g.. .. tt. '. -S5N"... Jl . >. l'. lt. 1tr/ L' .y-' '. ,. ,/f. le". ' 30 t. 1SO. t30. 1+O. iso- lca. Fig. 16b. 200-mb chart for 1200 GMT, Jung 26, 1959..

(38) 58. T. Masui. 4N. t1 11 Jl .,)･ 'N>tC".,fx. 11 t,./. W. /" - "!r. ':'- '7."-. -io,'. '. '. ,.< 4((.. ->. 'x "-'1. tt. '. .m---J. t'. '. u. [J. 12200. Il;ss (Ig(c. .. -o. ) ... ,tx'''. U,.(,--i7k-x- ?(5. `. ). )iNxx i,. ,7. <. ."--..-50 ,. '. r. Li. x,, (l. -･. 1 <xSIk, // .t c'""'K>xif. .-.. `,"". 170. t. S,' vu/i/ X fa. s. "ll- ---.'r. --s----. Xll; ice. S50. t -50. 7. t. -+s i3C "-.f, t+O rN. I2ot. blC. .,e. tt. t. Tx:,x,. 9. '. '. >･. ' t. '. l?. s/. Vl. H 1･'. ;. hse. L 3e. '. N. s. s. s. xtg,k".4"' "' <f/f. 3". tw. f r,,t' ' f,' .IY. t. N. N --. -55-----. s. N. N. s. N. N. (. s. H. 2o. 12400. x. 12 OO. . (-. .ko /r 13e. lga. 110. i30. b2e. 200-MB CHART 1200 GMT-JUN.27,ISSS. June 27, 1959.. L+O. tso. ･･. '. -S5J- -''. N. s. N. x. x N< s. N. ?' -s. H. ･}. --. -.NNSs. ･). N N. N. N. <･. s '-. N. N. N. , ek t' i. t k. i. -e. N. k. I. v. t6A-OO. .. N N. N. -". N N :. N. -. S5 s. N s s. ili. -. fl s. tr. Q. L '---.-. NXN 4svt30. s. -. "'. 3e--. l50. -"----'-". ----70. H 'x c. ". '･. ). po. 20. '. ). y" ). t. '' r･1 ('. o' ). --･AtD-.--cf. i40. / uN7. isd. ". IGGOO. -6S----"-'. .---70" i.. X-- ---- W. -GO----T -.---. "-F"'J'-""J''-- -""'-----. gNN. ..... v･･. 12 Ns. s. <v-..S"Uk,A･'. x iX... S7!. f. '. ... -60. < SNsN. JdN NN.. tIY'. tl. -( -. i k,. glf b/,f',l?}?Z,/&JN. .. N. ,(. NNi. ss. ..-v-. i?D. '75s. v. xQ ･ lfK J'. fo. IX (1. '- 1 ./lsS. t='N. N. K. N. -7.0 N. N. ". .. N N. s. J70. L. W. )(. J`,O. lgo. a. '. sN. ' '. 16e. I5. 1200 GMT,. Fig. 16c. 200-mb chart for. N. -55. 12+OO. N. 20. x. fi;N. N. sNx'. U'. -65. s. t30. Fig. 16d.. 16goo. lcgoo' o. 100-MB CHART l200 GMT-JUN.26,fS5S T5a. 100-mb chart for 1200 GMT, June 26, 1959.. lao.

(39) Summer High-Level Cyclones in the Region of Japan. g-12 5:. / --. O 8--L2 ... 1. p. 50. '---- "O-. /. 59. '. d L i'iltillllil.ll.e>lii2,. c(f) }o -. 6.'e-d9. Q. lo-ooo Og-l2. 3-. t{pi ;... l >a. ]o. 3o. e. f. 20. 20 130. l+o. ISO. Fig. 17. Track of the high-level cyclone, July 6-11, 1958?. on 200-mb charts. Other legend the same as Fig. 8a.. 9 it stagnated near a point 330N, 1130E, reaching its maximum depth. After this it began to move northeastwards with the trough retreating northwards･ At OOOO GMT, July 11, it reached South Hokkaido, then moving east-northeastwards. At 1200 GMT, on the same day, it was situated near 440N, 1480E. and then moved southwards all of a sudden. The cyclone reached a point about 40.50N, 1470E at 1200 GMT, July 12, and then moved east-northeastwards with the trough.. The eastward movement of the cyclone at the 200-mb continued while the high pressure region at the 100-mb was extending. With the stagnation. of this high,'the cyclone moved southwards and then northwards in the trough. It was moving southwards for a short time, while a northeastward extension of the anticyclone at the 100-mb took place again.. At the 500-mb an anticyclone extended from the'west and a trough persisted along the south coast of Japan during July 6rv8. After that, the anticyclone extended from the west over the Sea of Japan.. On the surface maps a front moved southwards from the Sea of Japan and in some region was found a slight rainfall affected by a weak cyclone. On July 9tNvlO a front was persistent. along the south coast of Japan and,. consequently, Japan was subjected to a weak "Northeasterly" situation. After July 11 the regular "Northeasterly" situation dominated over Japan.. ExAMpLE 5b. Southeastward-moving high-level cyclone south of Japan, July 10rv18, 1958. Type VI. Figs. 18aAv18b..

(40) 60 T. Masui. On the 200-mb chart a cyclone developed off the coast of San-in in a trough extending southwards on July 10. It moved southwestwards for a short period and then eastwards. At 1200 GMT, July 11, it reached a point near Shionomisaki, deepening and having a lower temperature part in its centre. After this its moving speed slowed down gradually by the effect of a typhoon moving northwards. It was situated at about 300N, 1410E at 1200 GMT, July 13, reaching its maximum depth. Meanwhile, the typhoon went away, at slow speed, towards the east. On July 14 the cyclone reached a point near 290N, 1420E and then moved east-northeastwards. At 1200 GMT, July 15, it was situated near 300N, 1420E and then recurved northwards. On July 16, it moved northwestwards by way of a point at about 310N, 141.50E. At 1200 GMT, July 17, it reached a point near 330N, 139.50E and then recurved towards the south, following a very erratic course. At this time a high temperature region was observed to the north of the easterlies. On July 18 the cyclone was displaced southwards in the neighbourhood of a point 31.50N, 1400E, after which it decayed moving perhaps northeastwards. Meanwhile, the intense Typhoon No. 5811 had progressed northwards and. ". '. hence the trough retreated northwards. Coming immediately after the situation of Example 5a, the anticyclone which had extended from North China retreated slightly at the 100-mb, while " the cyclone at the 200-mb developed. This'anticyclone extended a little to-. wards the east when the cyclone at the 200-mb began td move eastwards. The movement of the cyclone was carried under'the trough on the 100-mb ' '. no l-o lson z. -. -". .s- fe. ({). T･ 10-OO fZ-. I.. tttt'. <Dv 6i2. x. (} "te-ol,2. k .L ,'IS-12. hi6<e( vvx/ .10-12. /. t7-12. M IG-12. 5'3s'st2. 12-120 lg6Bis･i,.<l. 30 Xl/ 1+12S(2-12 ,3 i. d. f. S7S 5n-t2 9Gs. S l3-12 x zo xx S 12･12. 1. Nxx... 5Jl-12. ". 20. lS<%;?2. s6tt. ./ Fig. isa. Track of the high-level cycione, Jg,Iy.j,iQ,IE,is, lgss,t lse. .. t+e. on 200-mb charts. Other legend the same as Fig. 8a.. ..