UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The fine structure of the E-region Belrose, John Skelton 1951

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1952_A7 B3 F5.pdf [ 21.78MB ]
JSON: 831-1.0103255.json
JSON-LD: 831-1.0103255-ld.json
RDF/XML (Pretty): 831-1.0103255-rdf.xml
RDF/JSON: 831-1.0103255-rdf.json
Turtle: 831-1.0103255-turtle.txt
N-Triples: 831-1.0103255-rdf-ntriples.txt
Original Record: 831-1.0103255-source.json
Full Text

Full Text

THE FINE STRUCTURE OF THE E-REGION (original copy) THE FINE STRUCTURE OF THE --REGION by JOHN SKELTON BELROSE A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE i n the Department of ELECTRICAL ENGINEERING We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF APPLIED SCIENCE Members of the Department of ELECTRICAL ENGINEERING THE UNIVERSITY OF BRITISH COLUMBIA October, I95I i . THE FINE STRUCTURE OF THE &.RESION (abstract) Introduction Very l i t t l e i s known about the fine structure of the E-r©gion of the ionosphere. The pulse method devised by Breit and Tuve i s used to study the E-region i n detail. Observations were made at Vancouver for the months of July and August 1951* The frequency 1,5 to 5 mc/s* (down to ,5 mc/s, after midnight) was swept manually recording i n 100 kc/s. intervals the virtual height, h*, to the nearest kilometer,; Experimental (h 1,t) records were also taken at 15 minute intervals throughout the day on 2 mc/s. The (h*,f) curves were analysed for fine structure details of the region which are not recorded by ionospheric equipment used for routine obser-vations of the entire ionosphere. The following investigations were attempted j 1. Fine Structure of Night-Time E-Region. 2. Diurnal Variation of Fine Structure of E-Region. 5. Sunrise Effects of E-Region. 4. Occurrence of Echoes from Levels Below the E-Region. 5. Diurnal Variation of C r i t i c a l Penetration Frequency of E-Region. 6. Determination of Scale Height of the E-Region. Results 1. Fine Structure of Night-Tlme E-Region Throughout the night ionization generally appears as patches from random clouds. Near sunrise short-lived echoes are found between 80 and 200 kms. Few usable results showing fine structure details are found. i i . 2. Diurnal Variation of Fine Structure of --Region Experimental (h',f) curves are compared to derived curves for a simple parabolic region with an E 8 layer appearing as a sharp boundary embedded i n the simple region. Good f i t s to the theoretical curves are normally found below the cusp frequency. The H a i l ' appearing after the cusp frequency generally has a slope greater then that predicted. Various types of ledges found i n the --region are discussed. Moving ledges are often found with an approximate quasi-period (i.e. time to pass through the region) of half an hour. The variation of the penetration frequency of high smooth E s regions also appears to have a similar period. Once during the period of observation both these phenomena occurred to-gether. Very pronounced ledges are sometimes found above the normal maximum. 5. Sunrise Effects of E-Region Day-time ionization of the E-region commences before ground sunrise. Commencement time i s found to be approximately that time at which the sun*8 rays strike the E-region after grazing a spherical surface 59 kms. above the earth. 4. Occurrence df Echoes from Levels Below the E-Region Strong indications of region D are found. Often patches of ion-ization, as though from small ionic clouds, appear at various heights from 80 to 200 kms. No retardation effects are observed for any of these records. 5. Diurnal Variation of C r i t i c a l Penetration Frequency of E-Region The c r i t i c a l frequency of the E-region i s found to obey approximately a law i i i , . f c * kco8 n 7< where X is the sun's zenith angle. The average morning value for the index, n, i s .301 and the average afternoon value i s .35 . The average morning and afternoon value i s .325 . Examination of the (log t , log cos X ) curves show that the afternoon values usually f a l l more nearly i n a straight l i n e . 5. Determination of Scale Height of the E-Region a. Analysis of (h'.t) records From plots of the function constant -f- h/H H i s found to be 11.5 kms. b. Analysis of (h',f) records From plots of the function h' * hg+ </>(f/fc) Hi i s found to be 9.4 kms. When this i s corrected for a parabolic assumption giving the best f i t to a Chapman distribution, H » 11.28 kms. University of Bri t i s h Columbia October, 1951 I. TABLE OF CONTENTS I. INTRODUCTION: I II. EQUIPMENT . . . . 5 1, The Transmitter . . . . 5 2. The Antenna . . . . 5 5. The Receiver . . . . 5 4. Timing and Presentation Unit . . 4 III. MEASUREMENTS . . . . 5 Echo Splitting . . . . 5 IV. THEORETICAL CURVES FOR SIMPLE REGIONS . . 8 1. Reflection of Waves from the Ionosphere . 8 2. Distribution of Electron Density with Height . 10 5. Comparison with Experimental (_',f) Curves . 12 4, Theoretical (h1,t) Curves i n the Presence of a Sharp Reflecting Boundary . . . 15 V. FINE STRUCTURE OF NIGHT-TIME E-REGION . . 17 VI. DIURNAL VARIATION OF FINE STRUCTURE OF THE E-REGION 18' 1. Low Smooth E g Region . . . 18 2. Formation of a Low Smooth E f i Region . . 20 5. High Smooth E 8 Regions . . . 22 4. Rough E 8 Layers . . . . 25 5. Ledges . . . . 25 6. Ledges Near or Above the Normal Penetration Frequency 25 VII. SUNRISE EFFECTS OF REGION-E . . . 27 VIII. OCCURRENCE OF ECHOES BELOW THE E-REGION . 51 IX. THE DIURNAL VARIATION OF CRITICAL PENETRATION FREQUENCY OF REGION-E . . . . . 52 II, X. DETERMINATION OF SOALE HEIGHT OF THE E-REGION . 55 Comparison with Analysis of (h 1,f) Records . 57 ACKNOWLEDGMENTS . . . . 59 LITERATURE CITED 40 BIBLIOGRAPHY . . . . . 42 ILLUSTRATIONS . . . . . 46 III. ILLUSTRATIONS (follow page 45) FIGURE * 2, >Experimental (h',f) curves for August l4 and 16. 5.J 4. (a) Refraction of ray i n Ionosphere. (b) Parabolic Gradient of Electron Density. 5. Graph of the function <j) ( f / f c ) . 6. Theoretical (h*,f) Curves with a Sharp Increase of Electron Density Near i t s Maximum. 7. Typical Structure of Night-Time (Normal) Region-E. 8. Comparison, of Experimental (h 1,f) Curves with Derived Curves for a Simple Parabolic Structure with a Thin Sharp Layer Embedded in i t . 9. Variation of Penetration frequency of High Smooth E 8 Layers. 10, Formation of Layers above the Normal Maximum 11, Formation of Ledges i n Region-E 12, Occurrence of High Smooth E g region with high penetration frequency. 13, Occurrence of High Smooth E g region with low penetration frequency, 14, Occurrence of Low Smooth E s region with high penetration frequency. 15, Occurrence of Low Smooth E s region with low penetration frequency, 16, Variation of Penetration Frequency with Time for Month of July, 17, Diurnal Variation of the C r i t i c a l Penetration Frequency, 18, Graph of the Function, In-Jtyw constant + h / H i 19, Sample Plot of h 1 vs. (j) ( f / f c ) . I V F L A T S X a . The 250 K i l o m e t e r S o a l e b . One K i l o m e t e r M a r k e r s X I a . S p l i t F - e c h o b 0 N o r m a l E - e c h o a b o u t 125 kms. I l l a . S t r o n g E - e c h o a b o u t 139 kms. , a n d weak echo a b o u t 52 kms. b . S t r o n g E - e c h o a b o u t 130 k m s . , a n d weak e c h o e s a b o u t MS a n d 58 k m s . c . S t r o n g E - e c h o a b o u t 126 k m s . , a n d weak s c a t t e r 150, 182, a n d 212 k m s . I V a . E c h o a b o u t 85 kms. o n a f r e q u e n c y o f •14-75 m c / s . b . F r o n t v i e w o f t r a n s m i t t e r a n d r e c e i v e r s h o w i n g t r a c e s o n C . R . T , d i s p l a y . V I o n o s p h e r i c s t u d i e s , h u t s h o w i n g 85 f t . p o l e a n d d e l t a a n t e n n a s . LIST OF SYMBOLS A absorption constant of radiation c, velocity of light i n a vacuum. e charge of an electron-f frequency f c , or f°,# or f Q penetration frequency of ordinary component for E-region: f]j, or f x penetration frequency of extra-ordinary component for E-region cusp P e n e* r a^^ o n frequency for E-region % gyro-frequency for E-region H }E?_ s scale height mg h height h' virtual height height of lower boundary of a parabolic electron distribution hjj, or hjj height to maximum electron density of a parabolic electron F distribution (h 1 ,f ) virtual height vs. frequency record for a definite time (h 1 ,t) virtual height vs. time record for a fixed frequency I ions produced per unit volume per second In natural logarithm log logarithm to the base 10 m mass of an electron, except i n the expression f o r scale height where m i s the molecular mass of the a i r N number of ions per unit volume q x rate of ion production t time U group-velocity V p phase velocity y r distance i n kilometers (usually measured from height h Q) y 0 height from lower boundary, h Q, to sharp increase of electron density/ y m , or yjjp semi-thickness of a parabolic distribution z h -Hi —' frequency i n radians/sec, ^ refractive index of medium M- permeability constant £• dielectric constant /^v> refer to vacuum refer to region a function defined to be f/f, c 2 In: 1 - f f / f c _ ! 1 - f / f c coefficient of recombination coefficient of attachment S declination of sun ?C sun's zenith angle 0 depression of sun below horizon ^ latitude of observation f* molecular density a function tabulated by Chapman a secX ifX<85° VII. NOTE ON THE RECORDING OP VIRTUAL HEIGHTS For fine structure analysis we are interested i n changes in virtual height and not in actual virtual height above ground level. In recording the height of the D-region the author measured from the centre of the ground pulse, with the receiver gain reduced such that the ground pulse had an amplitude comparable to that of the echo, to the centre of the echo. This resulted in a subtraction of 15 kilometers from the height read off the scale since the centre of the ground pulse is at 15 kms. The equipment has a variable transmitter time delay control which consists of a monostable multivibrator that is triggered by 50 km. pulses. Hence i t is possible to shift the triggering of the transmitter f i r i n g i n 50 km. jumps. It i s noted that on certain positions of the transmitter f i r i n g i t is possible to exactly zero the centre of the transmitter pulse. On the range generally used the most stable point seems to be that for which the centre of the trigger pulse i s at 12-12.5 kms. The harmonics of this leak through the front-end of the receiver near 1,9 mc/s. This shifts to 15 kms. when the transmitter i s put on the a i r . Now the reason for this i s not clear. By shifting the transmitter delay control knob one shifts the bias and hence shift the position of the trigger pulse on the grid R-C decay voltage and the rise time of the f i r i n g pulse would shift the f i r i n g point of the multivibrator. However this would not account for the very large shift found. Also i t does not account for the non-uniformity of starting point as we jump back the delay i n 50 km. jumps. It appears from analysis of the record on the C.R.T. display that 15 kms. should be subtracted from a l l the readings, but from c i r c u i t analysis i t does not seem possible that shifts of this sort are obtainable. Until the reason for this phenomena i s found i t would be best to leave a l l records as recorded. Hence whenever reference i s made to the height of a layer i n this thesis i t is as read from the C.R.T. display. It i s noted that i f 15 kms. i s subtracted i t makes hjj s 120 kms. a value found by other workers. 1. I. INTRODUCTION Ionospheric characteristics are usually determined experimentally by measuring the amplitude and time delay of reflected signals as a func-tion of frequency. The commonest method used i s the pulse method origin-a l l y devised by Breit and Tuve (1926)t A sueession of pulses of short duration (JO to 100 u-sec.) i s sent ve r t i c a l l y upwards at regular inter-vals (1/50 to 1/120 sec.) by a transmitter. The receiver, which is located close to the transmitter, picks up both the direct and the reflected sig-nal. The spacing between these signals on a fast time base of a cathode ray oscilloscope gives a measurement of the height of the layer. The height so measured i s the vir t u a l height (h 1) of the layer and i s higher than the true height of the lower edge of the layer due to group retarda-tion of the pulse packet i n the ionized region. There are about 50 stations i n the world making routine ionospheric measurements. However, very l i t t l e i s known about the fine structure of the E-region. There are several reasons for this ( Straker (1950) )t (i) Since routine ionospheric measurements are made primarily with the object of providing data for forecasting radio propagation conditions, great precision i n measurement of h' i s not required. The obser-vations are made at hourly (sometimes 15 minute) intervals. The frequency range 1 - 15 mc/s. i s swept and the height range 0 - 1000 kms. i s displayed on the cathode ray tube display. ( i i ) Usually, no attempt i s made to make observations below about 1 - 1.5 mc/s. because the c r i t i c a l frequency of region-E i s generally found at frequencies greater than 1.5 mc/s by day. Also interference from broadcasting stations, and increased attenuation of the ionospheric * A l l references given i n "Literature Cited". wave as the frequency i s reduced obscure workable results. Observations were made i n Vancouver for two months (July and August), with a view to studying the E-region i n detail. The power output of the pulse transmitter i s 15 kw. The available frequency range i s .425 ** 5»® mc/s, A 50 u-see. pulse was used. The receiver has a sensitivity of 1 u-volt for 6 db. signal-to-noise ratio and a recovery time of approx-imately 10 u-sec. The antenna consisted of crossed deltas and i s 85 feet high at the vertex. The operating site i s entirely i n the clear. It i s situated on about the highest point on the campus of the University of Brit i s h Columbia. The approximate geographic location of the site i s 1 2 J . 2 J ° W. longitude, 4 9 . 2 7 ° H. latitude. The total magnetic f i e l d at - 4 ground level i s .575 x 10 webers/m . The gyro-frequeney for region-E (120 kms.) i s , thus, approximately 1520 kc/s. The scale measurement accuracy i s of the order of~t . 5 kms. The experimental recording of h' for the (h 1 ,f) curves i s about - 1 km. under quiet conditions. The freq-uency range i s manually swept recording h' i n 100 kc/s intervals from 1.5 to 4 mc/s. during the day and down to . 5 mc/s. after midnight. The following investigations were attemptedt (i) Pine Structure of Night-Time E-Region ( i i ) Diurnal Variation of Fine Structure of the E-Region ( i i i ) Sunrise Effects of Region-E (iv) Occurrence of Echoes from Levels Below the E-Region (v) Diurnal Variation of C r i t i c a l Penetration Frequency of the E-Region (vi) Determination of the Scale Height of the E-Region. II. EQUIPMENT 1. The Transmitter The transmitter was designed and built by the Radio Physics Labora-tory of the Defence Research Board. Necessary modification to cover the frequency range .5 mc/s to 5»8 mc/s was made by the author. Very poor L/C ratio was obtainable at the low frequencies. External clipping of 200 uufd. fixed capacitance across the oscillator tank resulted i n better wave-forms and made the low frequency limit .425 mc/s. The unit consists of a J kw. push-pull Colpitis oscillator, using a JE29 tube, driving a periodic push-pull class B linear amplifier using a pair of 715-B tubes. The peak out-put i s 15 kws. The pulse width i s 50 u-sec. (i.e. 7§- kms. on the time scale). 2. The Antenna The antenna i s a set of crossed-deltas. That i s two delta antennas, one for transmitting and one for receiving, are errected with a common vertex and the base line of one i s rotated through 90°. The vertex i s 85 feet high. The total base length i s 180 feet. The legs are 8 feet off the ground. The antennas are terminated i n about 800 ohms. 5. The Receiver The receiver was especially designed for the job by Niblock for the Radio Physics Laboratory of the Defence Research Board and constructed by them in Ottawa. Necessary modification to cover the range .5 mc/s. to 1.5 mc/s. was done by Niblock, now of this department. The band-width i s 21 kc/s. at 5 db. down, recovery time i s 10 u-sec, and sensitivity i s 1 u-volt for 6 db. signal-to-noise ratio. The tuning range i s now 200 kc/s to JO 4 . mc/s. Design notes and a more complete description of the unit i s given by Nibloek (1951) . 4. Timing and Presentation Units It was desired to be able to measure h 1 to an accuracy of about \ kilometer. Since 6 2/3 u-sec. pulses are required for 1 km. markers the scaling c i r c u i t i s controlled by a 75 kc/s crystal oscillator doubling to 150 kc/s. That i s i Blocking oscillator dividers were used because of their a b i l i t y to generate very narrow pulses. The stability of division rate can be made comp-arable to that of other relaxation oscillators, and once operating temp-erature i s reached relatively l i t t l e trouble i s experienced with division rate jumping. Good presentation of 10, 5°» and 200 km. markers i s obtained. The unit was b u i l t by Moore of this department. The marker presentation i s shown i n plate I. f a 1 t *» 150 kc/s. 5. III. MEASUREMENTS A l l (h!,f) records were observed on the 250 kilometer scale. That i s 250 kms. are spread out on one trace. The one kilometer markers are presented on this scale but seem to be masked by the two km. markers which seem to be stronger. This display i s inherent i n the vertical amplifiers of the C.R.T. unit. A l l readings of h* are subject to about i" 1 km. error for quiet conditions. The frequency 1.5 mc/s. (,5 mc/s. after midnight) to 4 - 5 mc/s. was swept manually and the value of h' recorded i n 100 kc/s. intervals. If any sharp jumps occured i n the record 50 kc/s, and sometimes 25 kc/s. increments were recorded. The time for the recording of one frequency sweep was about five to six minutes providing absorption was not too great so that the operator had to search for a weak echo i n random noise. A l l (h',t) records were observed on the 100 km. range. That i s 100 kms. are spread out on one trace. This can be expanded however so that 10 kms. are spread out over the face of the C.R.T, display. It i s quite easy to estimate to about \ km. The probable average error of recording an echo under quiet conditions is about •§• km. Measurements of (h',t) records on 2 mc/s. were recorded at 15-minute intervals from 0800 to 1600 P.S.T. for the period of June 18 to June 22, and July 5 to August 10, 195L Approximately three weeks of continuous half-hourly (fifteen-minute at sunrise and sunset periods) height-frequency records were recorded during the period of June 18 to August 17. Sunrise and sunset records were recorded throughout the whole period, except for weekends. Echo Splitting A very peculiar fading was observed for the echoes during conditions when high absorption was present, usually around noon, and during dis-6. turbed times at sunrise and sunset. The echo (as i n plate II) would become f a i r l y broad, then i t would s p l i t i n the centre and form two echoes from one to ten kms. apart. Usually one of the echoes would be stronger than the other. The weaker of the two echoes would then fade. The peak of the remaining echo would then be from £ to 5 kms. higher or lower than the original echo. This echo would then spread and f i n a l l y occupy approx-imately the original position. On the next fade the opposite component echo usually did the fading. The variation i n shape of the envelope can be explained by assuming the presence of several component pulses adding together with random relative phases to produce the observed result. The theoretical examination of the effect of the superposition of one pulse upon another with small time difference between them shows the result i s a single pulse as long as the two components are almost i n phase, but a change of amplitude or phase of either component produces a change i n the peak of the resultant. For example, where i t i s assumed that the pulse has the form y • 1-f- cos© and the separation of the two components i s a pulse-width, a change i n amplitude of one of the components from zero to unity (i.e. to equality with the other component) w i l l cause the peak of the resultant pulse to move \ of a pulse width. In this case where a pulse width on the time scale i s 7§- kms. this w i l l mean a sh i f t of 1 7/8 kms. On the other hand i f the two components should be out of phase and comp-ari t i v e l y close together, the resultant i s two pulses with a separation greater than the true separation ( Halliday (1956) ). Why there should be more than one reflected pulse i s more d i f f i c u l t to explain. Halliday (1956) found sudden changes i n (h 1,t) runs. He suggests that the two pulses causing these effects occur when the extra-ordinary component of the ray i s returned from region-E with an amplitude comparable with that of the ordinary component. It i s the superposition of these two pulses which causes jumps i n the (h 1,t) curves. Helliwell (1O4Q) suggests the possibility that the lower part of the E-region act-ually consists of a number of patchy nonuniform layers, each of which contains 'holes' large enough to permit some energy to penetrate (at nearly vertical incidence) to patches of ionization i n the next layer. Another explaination i s that the reflections of longer delay come from patches of ionization which are about the same height as region-E but not directly overhead. This i s probably what happens near sunrise. However at noon, since region-D seems to be strongest when this fade i s considerable, i t might be that region-D i s very patchy and part of the energy i n the pulse packet suffers partial reflection and partial transmission by a partly obscuring cloud i n region-D. The main portion of the energy i s unaffected. On reflection from region-E this energy retarded by the cloudlet i n D reach-es the receiver out of phase with the main pulse, thus causing the above effect. Examination of (h* ,t) records shows that the random variation of h' i s considerable. The mean position of the wandering echo was always recorded. 8. IV. THEORETICAL CURVES FOR SIMPLE REGIONS Before we can analyse a (h',f) curve for fine structure we need to know just what such a curve t e l l s us. Hence a very brief outline prepara-tory to the derivation of theoretical (h',f) curves for simple regions w i l l now be given. The affect of the earth* s magnetic f i e l d i s neglected. Whale (1950) pointed out that the neglect of the earth's magnetic f i e l d i s not permissible for (h 1,f ) curves recorded i n Cambridge. Since the magnetic f i e l d at Vancouver i s considerably higher than that at Cambridge (.575 gauss as opposed to .471 gauss for Cambridge) i t was thought that an analysis neglecting i t s affect on (h*,f ) curves for Vancouver would be insufficient. However i t i s seen later that a very good approximation for h' up to the penetration frequency results from considerations of a simple parabolic structure neglecting the effect of the earth's magnetic f i e l d . 1. Reflection of waves from the Ionosphere The ionosphere i s a dielectric region containing free electrons and ions. Under the influence of a passing electromagnetic wave the charges have imparted to them oscillatory motion that both absorbs and reradiates some of the energy. The ionosphere can thus be treated as a charge-free region having an effective dielectric constant and an effective conduct-i v i t y . The effective dielectric constant of the region is redueed below that of free space, and the region has an effective conductivity which depends upon the electron density and c o l l i s i o n frequency ( Jordan (1950) ). If the frequency i s sufficiently high so that the change In ionization density i s small i n the course of a wavelength , the ionosphere may be treated (by the method of ray optics) as a dielectric with a continuously variable refractive index. Under these conditions the wave penetrates the lower edge of the ionosphere without reflection, but within the Ion-9. osphere travels a curved path away from the region of greater electron density (i.e. smaller refractive index). By Snail's law (see figure 4-(a) ) sin</>i s /. ain0 or sin0 * sin 4>i (i) A The phase velocity of a region having negligible loss i s : v, a P ( i i ) where c • * « the velocity of light i n a vacuum, //•Vvtv J L r - permeability of region which i s unity for a l l non-ferromag-netic substances, £ r 8 8 effective dielectric constant of the region. Now i t i s easily shown that £ r « 1 - 47TU!©2  1 2 henee • - — 0 f i l l } 2 m-J Or another way of looking at this refraction i s that the wave front w i l l be twisted round because the phase velocity increases with height. If we define the ratio of the velocity of the electromagnetic wave i n free space to that i n the medium asyU. , the refractive index of the medium, then from equation ( i i i ) / • 2 " (*Ap> 2 - i - J L i ! dv ) The refractive index decreases as the wave front penetrates into regions of greater electron density. It i s clear from equation (i) that when^ 10, has decreased to the point whereJ^- «* sin0j', the angle of refraction w i l l be 90° and the wave w i l l be travelling horizontally. Hence at the top of the trajectory For vertical incidence &i » 0 or Lk n » © (v) Now the height that the wave-group w i l l travel i n an increment of time i s given by dh 3 U.dt, where U i s the group-velocity. In a dispersive medium of this sort, the group-velocity i s given by U«yUe (vi) so that dh • yUc.dt. But c.dt = dh', that i s , the increment of equivalent height over which the wave would have traveled had i t been propagated at the velocity of lig h t , so that *9^J- (Vil) Now h* i s the virtual height, and equation ( v i i ) states that the vir t u a l height depends upon the value of the refractive index along the actual wave-path. Therefore the refractive index, which depends on both freq-uency and electron density, makes h* dependent on both frequency and electron density. 2. Distribution of Electron Density with Height For the formation of a simple Chapman region R6 a N; 0 exp |\ - z - secX ®*P (-z) j ( v i i i ) where z 3 "* ^ ° H i H • kT/mg m the scale height of the region, 11. "X = sun's zenith angle, and N 0 i s a constant denoting the electron density at the level of maximum ionization forX " 0. N e changes with X j whenX " 0, i . e . when the sun i s i n the zenith, N e • N 0 exp £ £l - z - exp (-z) J .(lx) Over the region where z i s small, that i s , i n the region h Q H, we may write approximately 2 N e « Ho ( 1 - f " ) N 6 d - £ - 0 ) (x> where N e i s the electron density at any height y measured from the level of maximum electron density. From equation (x) we see that i n the neighbourhood of the level of maximum ionization the distribution of electron density with height i s parabolic (see figure 4-(b) ). L e t y m s ^ - h o " 2 H Hence from equation (x) >2 N e s Nmax 1 - (%___') T 'm (xi) or, i f heights y are measured from the lower 'boundary1 of the region, that i s from level hn N e = amax 2y - 7 (xii) Hence from equation (iv) 2 = i . . . V [_-_„ y m ym^  ( x i i i ) 12 where f ia the c r i t i c a l penetration frequency of the layer so that 2 i s equal to m f c . TT? The equivalent height h' reached by the wave i s thus given by h' s i dy /* y*0 = «o + vm £/£c 1 Q 1 + f / f c (xiv) 2 1 - f / f c After Booker and Seaton (lo4o), let us define a function </> (f/fc) * £/£c In 1 + f/f, 2 1 - f/f, c - 1 (xv) c so that equation (xiv) may be written h > a + y m ^ ( f / f c ) ...(xvi) The function ^ >(f/f c) is plotted i n figure 5. It i s now clear that having chosen a value for the scale height, H, i t i s possible to predict the shape of the (h',f) curve. 3. Comparison with Experimental (h* t f ) Curves From equation (xvi) we see that i f we plot h' a g a i n s t ^ ( f / f c ) the result i s a straight line whose slope is y m and which cuts the ordinate axis at h^. According to Booker and Seaton (1940) the parabolic law holds, for the region below the maximum, down to the level at which the electron density i s about twenty per cent of the maximum (i.e. f / f c • .447). Appleton and Beynon (194o) are, however, of the opinion that one should not use the (h',f) curves for frequencies below 0.9fc. Whale (1950) could not f i t simple parabolic curves like these to his data at a l l except for 15. frequencies very near the maximum. Whale also points out the well-known fact that practically any curve looks l i k e a parabola i f considered suff-ici e n t l y close to the maximum. He shows that the neglect of the earth's f i e l d for a l l high latitude locations i s not permissible. Nevertheless some 52 records selected from the month of July were examined neglecting the earth's magnetic f i e l d . These records were chosen as being those most free from pronounced ledges, and those that had the sharp increase of elect-ron density, so often found near the penetration frequency, above or very close to the height of the normal maximum^  Many records had low E_ regions s and therefore could not be examined using the simple theory just developed. These w i l l be discussed i n the next section. It i s found that most of the records examined can be f i t t e d to a parabolic distribution f a i r l y accurately down to f / f e equal to .58 (see figure 8), which i s the lowest frequency recorded for most records. The variation of scale height from record to record however i s considerable. The variation of scale height for July 17 i s given i n the following tablet July 17, O6I5 P.S.T. 0650 0645 0715 0745 I6I5 I650 1645 1700 1715 1750 17^ 5 * 7,65 kms. 4.55 6.5 9.55 8.16 12.9 U . 5 9.5 9.4 6.76 9.06 7.0 The average value for the month (some 52 records) i s 9.4 kms. 4. Theoretical (h'.f) Curves i n the Presence of a Sharp Reflecting Boundary In most of the experimental (h' ,f) curves taken there i s a ' t a i l ' at frequencies higher than the normal penetration frequency. It has been 14. suggested by other workers that this i s due to reflection from a sharp boundary situated near the height of the maximum of electron density of the simple layer. See figure 4-(b). The effective thickness of the region i s very small. This i s shown i n figure 11. For the record of July 16 at 1700 P.S.T. we have a sharp boundary occurring below the maximum, and the reflecting a b i l i t y i s such that the pulse penetrates this region with l i t t l e or no group retardation. The absence of group retardation i n penetration of this region i s shown since normal region-E showing through the sharp layer appears at approximately the expected virtual height. Appleton and Naismith (19^0) give calculations for an intensely ion-ized sheet embedded i n a simple 'parabolic' structure. Whale (1950) has extended the theory to include sharp boundaries above the maximum of the main region. Appleton and Naismith (19^0) assume a parabolic layer, and they neglect the earth's magnetic f i e l d . Whale (1950) shows that this i s not admissible for Cambridge. Vancouver has a stronger magnetic f i e l d than Cambridge. However, since reasonably good ' f i t s ' assuming simple regions are found for frequencies below the 'cusp' frequency i n cases where the sharp layer i s absent or above the normal maximum, i t i s thought worthwhile to carry the analysis further. It was shown previously that (x v i i i ) applies to the l e f t of the cusp frequency. The value of the cusp (xvii) and h! « h,, + y& f / f E l n x + f / f J o # ( x v i i i ) 2 ~ 1 - t/4 where fg i s written for the normal penetration-frequency of region-E i n order not to confuse i t with the cusp penetration frequency fgg. Equation 15-frequency; i s that for which the refractive index becomes zero at height y 0 . f° = f° BE *E 2yr 2 £ ,(xix) For frequencies higher than the cusp frequency = ho + 7 _ f / f E l a f fl - B T 1 l i n . ?o f j y_ f f ,..(xx) when y^ n y m equation (xx) reduces to equation ( x v i i i ) . The theoretical (h',f) curves for a scale height of 10 kms., after Whale (1950), are presented i n figure 6. The decimal fraction attached to each curve represents the c r i t i c a l frequency of the main layer to the frequency at which the sharp layer i s f i r s t reached both below and above the maximum. The (h 1 ,f) curve for normal region-E i s drawn so that the zero ordinate represents the height of maximum ionization density, h^ (i.e. f / f . a .834). It i s clear that even i f this simplified method gives worthwhile results the work involved i n f i t t i n g regions of varying scale heights to templates drawn for each new scale height i s tedious. Comparison of results to theory show that for most of the experimental curves examined the slope of the ' t a i l * i s too great for that predicted. See figure 8. Several curves, like the one for 0745 P.S.T. July 21, were found to follow very well the pattern predicted for the simple parabolic region with a sharply reflecting region at i t s maximum; However this curve appears to be unusual since most of the curves examined behave l i k e the one for 1715 P.S.T. August 15. The (h 1,f) curve for 1700 P.S.T. July 24 i s worth noting. The slope of the t a i l i s approximately that suggested for a sharp layer i n a normal 16*. regionaof scale height 10 kms. and ratio of frequency at which the sharp layer i s reached to normal c r i t i c a l frequency of .98 below the maximum. This reeord i s unusual since very few records with the sharp layer below the maxims—, were found which could be f i t t e d even approximately. * J1-" >•-. «' Whale (I950) found the same trouble. That i s -the slope of the t a i l i s far too great. He suggested adding a rectangular slab of electrons to the top ©f the parabolic approximation to give the correct slope for the t a i l . Whale shews that retardation through a slab of electrons gives the increase i n retardation near the penetration frequency that we need. How-we w i l l net try this method, since the number of possible combinations i s rather high. Manning (19^7) for the no-field ease, and Whale (1950) for the case where the earth's magnetic f i e l d i s not neglected, have developed a much better technique for direct analysis of experimental (h 1,f) curves. The process amounts t© inversion of the integral (xxi) i n which we have the observed h' as a function of f and we wish to obtain electron distribution as a function of height, However since the area under a derived curve must be found for each point on the true-height frequency curve the method is very laborious for the complicated (h',f) curves obtained for region-E. Whale used*a Brush differential analyser for analysis of his records. I t would be interesting to see how the (h',f) curves for Vancouver compare with those obtained for Cambridge, but lack of time prevents further studies of these curves. 17. V. FINE STRUCTURE OF NIGHT-TIME E-REGION Unfortunately night-time fine structure of the E-region was seldom observed. Records are of l i t t l e use for several reasons: Only after mid-night i s i t possible to explore the region below 1,5 mc/s. Even then the interference from broadcasting stations i s severe. Echoes from E are generally always found throughout the entire night, but are generally very patchy and too unstable to get anything out of manual frequency runs. Quite often a reasonable 'chunk' of ionization i s found, but i t i s always very thin, usually smooth and appears at random heights throughout the night. The only consistent thin layer of any extent i s found about 120 - 130 kms. Often during the early hours of the morning double layers about 20 to 50 kms, apart are seen over the whole frequency range. No retardation for these layers was ever observed. Some records showing typical structure of night-time E-region for a quiet morning are shown i n figure 7. Often records exhibiting retardation, l i k e that for 01^ 5 P.S.T. are found, but there does not seem to be any correlation from record to record l e t alone from day to day. I t i s believed by the author that looking for fine structure of night-time E-region: by manually sweeping the frequency is not possible. The changes i n the ion-osphere structure during a frequency run, about ten minutes i f the echoes are very weak, probably mask any pattern. Often the operator has to search for the weak echo i n random noise. Also with senders of high power many scattered echoes are always found. The records shown i n figure 7 show the f i r s t signs of day-time ionization commencing about 0330 P.S.T, with a penetration frequency of .85 mc/s. and a height of about 215 kms. The penetration frequency is seen to increase steadily, following the cos n X law, and i s about 2 mc/s. by ©500 P.S.T. Analysis for abnormal E was not attempted with available records. 18. VI. DIURNAL VARIATION OF PINE STRUCTURE OF THE E-REGION Many types of ledges are found i n the (h*,f ) records of region-E. The occurrence of different types of sharp layers and ledges observed at Vancouver for the period July to August 15, 1951 8 X 6 discussed. Whale (1950) has shown the d i f f i c u l t y experienced, short of complete analysis, i n dis-tinguishing between a ledge and a thin sharp layer with a very f l a t lower side (which we w i l l c a l l an E 8 region). Following the suggestion of Whale (1950) , we w i l l consider the region a ledge i f there i s appreciable re-tardation at the upper frequency limit. A detailed comparison with the observations of other workers i n this f i e l d i s also presented. 1. Low Smooth E 0 Region Frequently thin smooth layers are seen to move down through the normal region-E. These phenomena are usually accompanied by multiple penetration effects. The (h 1,f) curve may have a number of subsidiary maxima at various heights between 100 and 150 kms. As the thin layer moves down the region l i t t l e or no group-retardation i s found near the normal penetration frequency of region-E. Frequently there is a range of frequencies for which simultaneous reflections are obtained from region-E and region-F. Usually the height measurements were made on a scale range for which F could not be observed without switching ranges, hence the start of F was not generally recorded. One interesting set of records i s shown in figures, 1 to 5 for August 14. At 1530 and 1600 P.S.T. both normal E and F are clearly visible through the thin layer. The records for this day are also peculiar, since at I630 P.S.T. the E region suddenly vanished and the s records for I63O and 1700 P.S.T. show smooth (h'.,f) curves of normal region E which are entirely predictable from simple parabolic structure theory. From 1715 u n t i l 1900 P.S.T., after which the penetration frequency of normal region-E disappeared below 1.5 mc/s.) the usual ' t a i l ' again 10 appeared with region-F showing through i t . From this set of records i t would appears (1) that the ionization of normal region-E i s so increased by the presence of the sharp layer embedded i n i t that complete obscuration of f i r s t F-echoes i s possible. That i s , only after the sharp layer has disappeared or moved higher than normal region-E's maximum i s i t possible to observe both E (or E ) and F-echoes. (2) The rate at which the E.-region dispersed suggests that the ionic cloud producing this effect had drifted by (horizontally) between 1600 and 1630 P.S.T., unless sufficiently rapid diffusion of the thin layer i s possible. It i s noted that Whale (1950) also wondered whether or not a sharp layer would quickly disappear by diffusion of the electrons forming i t . There are two theories explaining the phenomena of low E a-regionst (1) "Scattering Cloud" theory suggested by Best, Farmer and Ratcliffe (1938), Appleton, Naismith and Ingram (1959), Appleton and Naismith (1959 and 19^0). This theory postulates that clouds of high ionization density are present i n region-E which are capable of producing scattered reflections on freq-uencies above the normal c r i t i c a l frequency. It has been found that with senders of high power, as i n our case, these echoes appear frequently both day and night on frequencies between 5 and 4 mc/s. The "overlap" between region-E and region-F i s explained on this theory by region-F being seen through "holes" between ionic clouds. Hence i t i s possible for suppression of F-echoes when the clouds become sufficiently intense for the spaces to be more or less f i l l e d up. The frequency on which echoes from region-F f i r s t appear i s interpreted as that frequency which can f i r s t penetrate the least dense portions of the patchy layer. The frequency on which reflections from E f i n a l l y disappear i s that frequency which i s capable of penetrating the more densely ionized portions. This theory has been developed by Booker (1950), who has suggested that the frequency i n which the echoes from E disappear may be related to the size of the ionic clouds rather than to their maximum density. 20, (2) "Thin-layer" theory. This seeks to explain the fact that simultaneous reflections may be obtained from region.E and F by invoking partial re f l e c t -ion and partial transmission effects which occur when the refractive index of a medium changes appreciably i n a distance comparable with a wave length. Under these circumstances! the reflection may be treated i n the same manner: as the refleetiomof waves at the surface of a dielectric that may or may not. have loss. The theory of this i s discussed i n Jordan (1950). Kirby and Judson (1955) suggest that the required sharp gradient of refractive index might occur at the base of the normal E. Since this suggestion does not agree with experimental results, later workers have assumed that an extra layer of ionization exists near the maximum of the normal region. I f this layer i s thin, the presence of the simultaneous reflections from E and F and the absence of group-retardation at the c r i t i c a l frequency of abnormal E can be explained. Findlay (1950) and Briggs (1951) have found the thickness to be a few hundred meters. Whale (1950) favours the thin layer type. This appears to be the most probable type for Vancouver, There are times how-ever when the 'overlap' i s large and pronounced irregularities are present. These records favour the hypothesis of scattering clouds. 2. Formation of a Low Smooth E 0-Region On August 14 (figures 1 to 5) we have a set of very interesting results. At 0950 P.S.T. we have normal region-E with a high smooth E_-region (to be discussed i n the next section) having a penetration frequency above 5 mc/s. At 1000 P.S.T. we note a very marked flattening and rounding of the group-retardation curve. By 1050 P.S.T, the (h',f) retardation has become comp-letely 'squared' off. The small t a i l has a low penetration frequency. By 1100 P.S.T. the record i s again normal except there i s a marked reduction i n group-retardation and the penetration frequency of the E 8 region i s very low. By 1200 P.S.T. there i s a complete 'blanket' of ionization with only a few irregularities near the normal penetration frequency of E. This 2 1 . continues u n t i l 1500 P.S.T. at which time the (h 1,f) curve s p l i t above the normal penetration frequency, and a second higher layer was vi s i b l e with a high penetration frequency. Neither layer shows any marked retardation effects. By I55O P.S.T. a definite layer i s visible above the main (h»,f) curve. This has grown i n extent and become much thicker by 1600 P.S.T. Then at 1650 P.S.T. an interesting (h',f) curve i s recorded. For two records we have complete absence of Eg layer effects and have ordinary normal E-region curves. This phenomena was previously discussed. From 1715 P.S.T. un t i l disappearance of normal E into the broadcast band (I9I5 P.S.T.) the records are again normal. That i s they show the E-region with an E a layer near i t s maximum. The more common formation and disappearance for low smooth E g layers seems to be as follows 1 The penetration frequency of the high smooth E 8 t a i l i s seen to increase to a high value about midmorning. A gradual flattening of the retardation i s observed u n t i l around noon the penetration frequency i s hardly discernable. This suggests the slow passage of a sharp E 8 layer down through normal E-region as has been postulated. About mid afternoon we see small group-retardation effects above the normal maximum and gradually the layer commences to get patchy. Small portions of normal E and F start showing through the thin layer. Finally the region i s again normal. The low E g layer seems to have dispersed. Usually a high smooth E g layer appears simultaneously with observance of normal E. The occurrence of low smooth Efl layers with high penetration frequency (i . e . above 4 mc/s.) and low smooth E 8 layers with low penetration frequency (i.e. below 4 mc/s.) are shown i n figures l 4 and 15 . After Whale ( 1950) , a black square indicates that the phenomenon i n question did occur during that half-hour, a white square indicates that the phenomenon did not occur during that half-hour, while a white square with a diagonal indicates that no record was taken during that half-hour. The heavy vertical lines indicate times of sunrise and sunset. The early morning records, i.e. before approximately 22. six o'clock, normally appear here as a white square. This does not always mean that the phenomenon did not occur since frequently the records were uunusable due to random variation of reflecting height throughout the slow process of manually sweeping the frequency. 5. High Smooth E f l Regions The smooth E_ regions discussed i n the last section occur below the maximum of the normal E-region. If this type of region occurs near and above the E-region maximum we shall c a l l i t a high smooth E 8 region. This type of sporatic E ionization i s of very common occurrence. An interesting feature often found is the manner in which the inaximum penetration varies. A set of records for July 10 showing a high smooth E region near or above 8 the niaximum of the normal E-region i s shown in figure 9. At 0815 P.S.T. the region appears as a very thin sharp layer with high penetration frequency. It i s located quite close to the normal maximum. By 0915 P.S.T, the pen-etration frequency of the E 8 region has shrunk to 5.8 mc/s. Half an hour later the penetration frequency of the thin layer has again increased to a very high value, and the thin layer has moved down the region. Again at 1115 P.S.T, the penetration frequency starts to shrink. It i s s t i l l however greater than 4 mc/s. but can now be recorded since i t i s within our avail-able tuning range. By 1245 P.S.T. the penetration frequency has again Jeducedeto about 5*85 mc/s. Half an hour later the penetration frequency has again increased to a value greater than that recorded. It i s inter-esting to note that the quasi-period, i . e . time to shrink below 4 mc/s, and to increase to a very large frequency, i s of the order of half an hour. Unfortunately since records were taken at half hour intervals this figure for the quasi-period i s only approximate, but whenever this phenomena appeared i t seemed to have a similar period. From 1515 P.S.T. on the E_ layer grad-ually f a l l s i n height and continues to move down through the region un t i l 2?. by 1415 P.S.T. the group-retardation cusp i s indistinguishable making further observations impossible. It i s interesting to note that Best, Farmer and Ratcliffe observed the same result i n 1938. They called this their "b-threshold frequency.™ Whale (1950) also recorded similar results. High smooth Eg regions occur very frequently i n region-E at Vancouver. However unfortunately they seem to form during f u l l day-light hours and usually records were taken on half-hour intervals. Hence the changes caus-ing the formation and decay of such regions have not been observed. The occurrence of high smooth E g layers with high and with low penet-ration frequency for the months of July and August are shown in figures 12 and 13» 4 . Rough E 0 Layers Quite often low rough obscuring patches of ionization are found. These are usually about 100 to 110 kms. During the day time these patches are usually of very short duration. Patches of ionization from such rough layers are often observed i n the early morning hours. Seldom are they completely obscuring however. Mostly this type of region i s very patchy and often i n the early morning hours i s present with other clouds of ionization at levels from 80 to 110 kms. making manual recording of (h',f) curves very d i f f i c u l t and often impossible. A l l records when low rough E layers are present indicate severe scattering. Dieminger (1951) also found similar short-lived echoes near sunrise. He suggests that these echoes are caused by meteors producing short-lived ionized t r a i l s i n region-E. 5. Ledges Small subsidiary maxima are often recorded i n the (h*,f) curves below the main maximum of the normal region. These small ledges do not appear for very long and often do not exist from record to record (fifteen minute 24. intervals). Sometimes they may persist for about an hour and the progress of the ledge i s always down i n height. Generally these moving ledges are found i n the afternoon. A set of records showing ledges of this type moving down through the main region are shown i n figure 11. The quasi-period at which they pass through the region i s usually of the order of half an hour. Whale (1950) also found a similar period. Examining figure 11, we see that i n the afternoon of July 6 a very broad ledge i s formed at 1745 P.S.T. This ledge has become much sharper by 1815 P.S.T. and i s moving down in frequency at about 10 kc/s. per minute. Half an hour later the ledge has gone. Similar results are shown for the morning of July 16 and morning and afternoon of July 17. The mean quasi-period at which the ledges pass through region-E i s approximately half an hour for a l l records shown, however the rate of decrease of frequency with 1 time i s widely variant since i t depends on the slope of the main (h 1,f) curve at the time. It i s interesting to note here that Briggs (195*) found similar phenomena by analysis of pamoramic records (i.e. (R,f) records) and he compares his: results to the (h',f) analysis of Whale (1950). The analysis of (h',f) records for Vancouver seem to be roughly i n agreement with the observations recorded by both Whale (1950) and Briggs ( I 9 5 I ) . Martyn (1950) has suggested that the apparent movement of a perturbation down the (h',f) curve is not real, but due to a disturbance travelling horizon-t a l l y with a t i l t e d wave front. It may be that effects i n region-E are similar i n nature to those discussed by Martyn (195°) Munro (1950) for region-F, Whale (1950) does not think that these effects could be caused by moving vertical stratifications in the ionosphere since the wavelength of the disturbance i n the ionosphere would have to be greater than 5 kms. for any effect other than a broadening of the echo to be found. A very interesting phenomena i s seen i n figure 9. The quasi-period for variation i n penetration frequency of a high smooth E layer was seen 25 to be approximately between •§• to 5/4 of an hour. Unfortunately whenever such phenomena occurred records were taken at half hourly intervals. On July 10 (see figure 9) at 0815 to 0915 P.S.T. a set of moving ledges i s seen to move down through region-E, i n the presence of variation of penetra-tion frequency of the high smooth E f l layer. Both these phenomena have approx-imately the same quasi-period. The similarity of the quasi-period of these phenomena was noted by Whale (1950)» n e never found them occurring to-gether. 6. Ledges near or above the Normal Penetration Frequency On several occasions a very distinct retardation (h J,f) curve was found for frequencies well above the normal penetration frequency. This phenomena is usually short-lived and seldom lasts as long as an hour. Several records of two occurrences for July 17 and 24 are shown i n figure 10. The second retardation i s unusual since i t i s too high i n frequency to be the penetration frequency of the extra-ordinary component. For Vancouver the total magnetic f i e l d at ground level i s .575 gauss (authority Carnegie Institution on Earth's magnetic f i e l d for 1947). Assuming that this f i e l d f a l l s off with the cube of the distance from the earth's centre (for a simple dipole) the calculated gyro-frequency at 120 kms. is found to be 1520 kc/s. With a magnetic f i e l d present the electromagnetic wave i s s p l i t into two components which penetrate the region at different frequencies. It can be shown that the gyro-frequency i s related to these frequencies by the expression where f x - penetration frequency of extra-ordinary component, f s penetration frequency of ordinary component. 26. Whenever such phenomena exhibiting double maxima were found i t was arb i t r a r i l y assumed to be the result of a magneto-ionic s p l i t . The gyro-frequency was calculated on this assumption for some eleven records. The values for the gyro-frequency were found to range from 1.4l to 2.01 mc/s. with an average of 1.71 mc/s. This i s ridiculous since the gyro—frequency at the pole i s only 1.6 mc/s. Hence this t a i l after the normal penetration must be a very pronounced ledge having appreciable thickness. A very thin layer must be situated near the penetration frequency of this pronounced ledge. 27. VII. SUNRISE EFFECTS OF REGION-E Ionized layers are formed by the absorption of ultra—violet radiation from the sun, the different layers being formed due to the absorption of different ionizing wavelengths by different atmospheric constituents. Following Chapman's treatment for the formation of a single layer i t may be shown that at the level of maximum ion production the number of electrons produced per unit volume q x i s H" " <lx " q 0 cos X (i) where q Q *> rate of production of electrons when the sun is overhead. However when ions, or free electrons, are formed they tend to unite with ions of the opposite sign to form neutral molecules or atoms. The rate of change of ion density i s hence given by the equation |f - q _ - * N 2 (u) where N s number of ions of one sign (or ion pairs), q x s rate of maximum ion production, <K « coefficient of recombination. Now at sunrise, i f we assume that a l l the ionization has been lost during the night, according to the simple Chapman theory layer ionization would not commence u n t i l layer sunrise. Hence we should expect ionization to commence before ground sunrise. D i f f i c u l t i e s were experienced i n locating f for very early hours of c the morning. After f i r s t signs of ionization were found the ionization seemed to be more i n patches then in layers. Also f_ was usually s t i l l below 1.5 mc/s. when the local broadcasting stations commenced transmission Since on very few mornings was i t possible to trace f c continuously from f i r s t signs of ionization through to ground sunrise, a l l readings for the 28 month of July were averaged i n sets of three consecutive days and then the results were plotted i n figure 16, after making the times of ground sunrise coincident with the time of ground sunrise for the middle of the month. The dispersion of points i s very bad. Since ionization began about one hour before ground sunrise, l e t us investigate the hour of sunrise at different atmospheric levels to deter-mine i f this i s sunrise time for the E-region. S o l a r rty$ t T T T T T The shadow of the earth i s of cyclindrioal shape. It i s clear that the height H at which the cylinder cuts the zenith i s given by His a ( - 1 ) cos 0 where a a radius of earth, 0 <» angle of depression of sun below horizon. ( i i i ) The hour angle h of the sun at i t s ri s i n g at height H i s cos J( s sin0 8 i n ^ -+~ cos<£ cos^ cos h where X » zenith distance of the sun, £ s i t s declination, (f) » latitude of the place of observation. Now X » 9 0 ° + & (iv) Taking into account 341 for horizontal refraction and 16* for semi-diameter (v) Mean time t of sunrise at a height H i s 29 t a 12 + £ _ h (vi) where £ a equation of time (Nautical Almanac), 12" a apparent time at actual noon. Prom figure 16 for the month of July 1951, h s 12.55 - 5.17 » 9.16 hrs. » 157.5° c o s X • sin ^9.27°. sin 21.57° - J - cos 49,07°. C 0 8 21.57° . c o s 157.5* X - 99.9° 9 - 99.9° - 90.85° « 9.07° and H- = 6560 ( 1 * 1 ) - 76.5 kms. cos 9.07° Since this i s considerably below the height of region-E we w i l l investigate the hour at which solar rays strike a particular atmospheric level by grazing a surface concentric with the earth. ft » From -the figure above i t i s obvious that when the suns rays strike B after grazing a surface d kms. above the earth, i t i s sunrise at A. By similar triangles OA - OB OM ~ ON or x r a ( SjLi*^ (vii) a + d 50. where BL S H AL & x M s d 0L = a = radius of earth. Now i f H s 120 kms. (region-E) 7«.5 = «5«0 ( 6 ^ 5 i ) d s 39 kms. Hence i t i s seen that the ionization of region-E begins to increase when the solar rays strike i t not by grazing the surface of the earth, but by grazing a spherical surface concentric with the earth and at a height of 39 kms. from i t . This i s just above the ozonosphere. This same result was found by Ghosh (1958) for Calcutta. The fact that agreement i s found i s interesting, but not conclusive because of the erratic variation of the penetration frequency during the hours before ground sunrise. The results do at least i l l u s t r a t e that ionization does occur before ground sunrise. The experimental method used for examination of sunrise effects was recognized as inadequate, since for manual frequency sweeps the change i n structure of the ionosphere i s quite appreciable over the length of time required for one record. Also even after midnight the interference from broadcasting stations i n Vancouver i s quite bad, often masking what appears to be the start of a retardation curve. Also as mentioned previously reflections from random patches about the height of region-E are found throughout most of the night. To overcome thiB d i f f i c u l t y (h',t) records were taken for one week at five minute intervals, near sunrise, on several frequencies. However broadcast interference masked the results on the lower frequencies before any trend of the curve could be found i n order to determine i f the echoes were coming from formation of day-time region-E or merely cloudlets. 51. VIII. OOOURRENOE OF ECHOES BELOW THE E-RESIOH Very consistent echoes were found from a region 58 kms. These echoes were usually strongest near 1,9 mc/s., but often were recorded over the whole frequency range. Various other echoes at 60 kms., 80 kms., and 100 kms. were often recorded i n patches of ionization. A l l regions were very thin. Often in the early morning the echoes at 58 kms. would be very uniform over the whole of the frequency range. A number of (h 1,f) records were plotted, but no trend whatsoever was found. Photographs of these echoes and a discussion of retardation effects on (h 1,t) records i s given Niblock (1951). 32. IX. THE DIURNAL VARIATION OF CRITICAL PENETRATION FREQUENCY OF REGION-E In a l l early measurements of the penetration frequency no clear dis-tinction was made between normal and abnormal E-layer ionization. Appleton and Naismith (1935) showed that the cusp of the (h 1 ,f) curve should be regarded as approximating the true E-layer c r i t i c a l frequency. The variat-ions of the penetration frequency of the E-region throughout the day has been studied by many workers, amongst whom are Best, Farmer, and Ratcliffe (1958), Tremellen and Cox (1947), and Whale (1950). The normal variation of ionization calculated from Chapman's theoretical analysis on the ionizing effect of monochromatic solar radiation on the earth's atmosphere leads to the result that N e * constant x (cos 'X ) (i) where N = maximum electron density, 7^  a sun's zenith angle. but f c << v£ _ hence f s k (cosV ( i i ) where f i s the normal penetration frequency. c Half hourly (h',f) records (quarter hourly near sun-rise and sunset) were taken over the period of about a month and a half. The variation of o x fg was observed since the extra-ordinary component f_ was never observed. On very few days of the period of observation was i t possible to trace the penetration frequency continuously throughout the day. The averaged values for morning and afternoon for three consecutive days was compared for solar control using the zenith angle of the sun for the middle day. A l l values for f°, used showed reasonably distinct cusps. The value of fj-° rises before noon, and reaches a broad maximum at noon, and f a l l s again i n the afternoon. (log f Q , log cos^C ) graphs were plotted and from the slope of the best 55 straight line through the points a value was obtained for the index, n, in the relation f c a k cos 1 1 X ( i i i ) The values found are given i n the following table. The index of the relation f » k cos n Date Morning Afternoon July 9 .511 12 .292 .312 17 ..315 .572 20 .28 .425 24 .5^2 .352 August 16 .277 .329 The average morning value i s .301 and the average afternoon value is .35. The average morning and afternoon value i s .325. Contrary to the results of Whale (1950) the average afternoon value i s higher than the average morning value. The afternoon values usually f e l l more nearly i n a straight lin e . The very early morning values were a l l considerably higher in frequency then predicted by the above analysis. This i s what would be expected since ionization begins before actual layer sunrise. Chapman (1931b) discusses grazing incidence conditions but since the E-region does not seem to be a Chapman region the necessary complications of introducing these conditions into the analysis was not included. A l l graphs seem: to follow a very definite trend, but the day to day variation i s quite appreciable. It i s believed by the author that i f observations were taken over a longer period of time a better average result could be found. The results seem surprisingly good considering the complexity of the (h*,f) curves. A sample plot of (h*, log cos 70 curves are shown i n figure 17* Some of the d i f f i c u l t i e s found i n determining f°, from the (h',f) records are discussed i n what follows. Days of severe absorption can be detected because the part of the (h 1,f) curve where h' increases rapidly with f i s i 3*. then blotted out. It i s very interesting to observe that echoes from 57 kms. are very strong at such times. In determining the c r i t i c a l -frequency for region-E another complication i s the presence of abnormal region-E phenomenon. There i s no very marked increase of h' with f . There are many examples during the afternoon and early evening hours (as i n figures 1 to 3) when the (h',f) curve has no marked group-retardation phenomena as f approaches f°.. Instead, the value of h' after reaching a certain value f a l l s rapidly to a height which i s maintained constant to the upper frequency of observation (i.e. 5.8 mc/s.). These low smooth E g-regions were discussed previously. Nearly a l l records showed the extension of ref l e c -o tion beyond fg at a f a i r l y constant le v e l . However, some were irregular i n nature as might be caused by scattered reflections from patches of ion-ization. On only a few records was there a sharp cessation of echoes beyond the normal H a i l * generally following the penetration frequency, A few records showing this are found i n figure 9 for 0845, 0915, 1115, 1145, and 1245 P.S.T. on July 10, Generally the records just trailed off into a patchy region. Occasionally when the height of abnormal E i s very low the o phenomena indicating fg are entirely absent and reflections from a constant level occur over the whole frequency range. Generally when this occured during the evening 28 mc/s 1short-skip * was present indicating the intensity of ionization has been increased by the penetration of abnormal into normal region-E. On many occasions echoes of weak intensity are returned from a level 90 to 105 kms. persisting throughout the whole frequency range. Generally normal region-E i s present and apparently unaffected. 55. X. DETERMINATION OF SCALE HEIGHT OF THE E-REGION: If the frequency under observation i s well below the normal c r i t i c a l frequency for region-E i t i s f i r s t natural to assume that reflection occurs i n the lower portions of the ionized region of the type discussed by Chapman (I95I a,b). He has shown that i f the ionization i s produced i n a homogeneous isothermal atmosphere by a monochromatic incident radiation absorbed according to a mass-absorption law, then the number of electrons produced per unit volume per second i s given by jexp £ -h/Hl- A Eftftyex? (-h/H) j ^ (i) where I = ions produced per unit volume per second, h • height, ft* » molecular density at h •» 0, Hi- - 'scale height 1 of atmosphere of molecular weight M at temperature T, A <* absorption constant of radiation, X - sun's zenith angle, "^CX}~ a function tabulated by Chapman, a e c * i f X < 8 5 ° . Following the method Budden, Ratcliffe, and Wilkes (1939) i t can be shown} assuming the reflection occurs at a considerable distance below the ionization maximum, and that electrons are lost by a process of recombination or attachment sufficiently rapid for the electron density (N) to be i n 2 equilibrium with the ionizing agency, so that «<N = I or ySN s I according as recombination or attachment i s assumed, a n d ^ and H are constant; the height h of any portion of the layer having a fixed value of N i s governed by the expression lnTOOs constant -f- h/H ( i i ) 56. which shows that h plotted against InjW) should give a straight line from whose slope Hi may be obtained. This equation for the E-region was also discussed by Appleton and Weekes (1939). The period at sunrise and sunset was not considered i n the above analysis since for very large values of sec j( (corresponding to grazing incidence of the beam of radiation) the level surfaces traversed by the beam can no longer be treated as parallel planes and Chapman's simple law developed above does not hold. Chapman (1931^) discusses grazing incidence but since virtual height only, not actual height, was plotted the complications at sunrise and sunset were not included since retardation of the echo makes these readings useless unless they are converted to actual height. The: virtual height during f u l l day-light i s approximately equal to the actual height up to about 80 per cent of the penetration frequency. On very few days did the equivalent heights follow exactly the simple law predicted by equation ( i i ) . However even though the random variation seems to be periodic no consistent trend seems to be followed from day/ to day. Hence the 'best' straight line was drawn through the dispersion of points. The morning and afternoon values for each day were plotted separately. Sample plots of h 1 vs. lxr|iH)for July 13, 16 and 17 are shown i n figure 18. The average morning value of the 'scale height', H, from the 'best' straight line for the (h', ln^ OO) curves i s 11.1 kms. The average after-noon value i s 11.5 kms. The average morning and afternoon value i s 11.3 kms. The results are shown in the following tablet Determination of Scale Height from (h' ,t) Records Date Morning Afternoon July 13 11.3 12.6 16 11.3 15.5 17 12.9 14.5 18 12.1 14.5 19 15.2 4.2 20 9.0 21 11.5 14.5 57. July 23 9.9 11.2 24 11.3 11.0 25 10.9 15.0 26 10.5 12.7 27 15.8 7.0 30 10.6 10.6 31 12.4 11.5 August 1 11.0 2 11.6 77 10.3 8 9.8 9 7.67 Comparison with Analysis of (h'.f) Records As discussed previously (h 1, (^)(f/f c) ) curves were drawn for a l l records i n which normal region-E was found. The average value of the scale height for the month of July i s 9.4 kms. Sample plots of this function are shown i n figure 19* These plots were based on the theory of a parabolic distribution of ionization. Pierce (1947) calculated the true height from virtual height records for a Chapman region, and compares his results with the results for a parabolic distribution. The parabolic distribution neglects retardation of the region through which the wave has just passed. Hence i n order to make a f a i r comparison of the value of H calculated from (h*,t) records with that calculated from (h',f) records both should be based on the same assumed electron distribution. Therefore after Pierce (19^7) rewrite equation (xvi) of Section IV. h* s \ + ymp <J>(*/*o> C 1 1 1 ) where h j ^ = height of maximum electron density for parabolic structure, ymp 5 semi-thickness of region for parabolic structure. The necessary corrections to be applied to the parabolic results for the best f i t to a Chapman distribution aret H » 0.6 y ( i v) 58 \ - % - 0.14 H: (v) Apply equation (iv) to the results of section IV. Hi- 0.6(9.4)(2) » 11.28 kms. It i s interesting to note that the value for the scale height, H, calculated on an assumed simple Chapman region from (h',t) records over the same period gave a value of 11.J kms. The average value for H calculated on an assumed Chapman distribution of some 52 (h*,f) records i s 11.28 Kms. The correlation between the two methods i s very close. 59* ACKNOWLEDGMENTS The author wishes to express his appreciation for the co-operation and financial support of the Radio Physics Labority, Telecommunications Establishment, Defence Research Board, Ottawa, who made this investigation possible. Thanks are also due Dr. T.W. Straker, R.P.L. Ottawa, for suggesting the subject, and to Mr. J.W.C. Scott, R.P.L. Ottawa, for helpful suggestions while he was i n Vancouver. Thanks are also extended to Dr. F. Noakes, grantee, under whose direction the investigation was made. Credit and thanks are also due D. Moore, student-assistant under D.R.B. Grant 242-245, and P.A. Niblock, colleague working under D.R.B. Grant 242, for numerous hours spent i n assembling and modifying the equip-ment and assisting i n the recording of results. 40 LITERATURE CITED Section I Breit, G. and Tuve, M., Straker, T.W., Section II Niblock, P.A., Section III Halliday, E.G., Helliwell, R.A., Section VI Appleton, E.V. and Naismith, R., Appleton, E.V. and Naismith, R., Appleton, E.V., Naismith, R., and Ingram, L.J., Best, J.E., Farmer, F.T., and Ratcliffe, J.A., Booker, H.G., BriggB, B.H., Phy. Rev., 28, 554 (1926) Personnel correspondence, November 1950 (1950) M.A.Sc. Thesis University of Bri t i s h Columbia (1951) Proc. Phy. Soc., 48, 421 (1956) Proc. I.R.E., 52* 887 (19^9) (1940) (1940) (1940) (1950) (19*7) (1950) Nature, l4 j i , 245 (1959) Loc. c i t . (1940) Proc. Phy. Soc., 5 1 , 81 (1959) Proc. Roy. Soc. A., 164, 96 (1958) Conference on Ionospheric Physics, Penn. State C o l l . (Quoted from Briggs (1951) ) (1950) Proc. Phy. Soc. B., 64, 255 (1951) Section IV Appleton, E.V. and Beynon, W.J.G., Proc. Phy . Soc., ^ 2 , 5I8 Appleton, E.V. and Naismith, R., Proc. Phy. Soc., ^ 2 , 402 Booker, H.G. and Seaton, S.L., Phys. Rev., 5J_, 87 Jordan, E.C., "Electromagnetic Waves and Radiating Systems," p. 665 Manning, L.A., Proe. I.R.E., 22, 1205 Whale, H.A., Dissertation submitted for Fellowship Trinity College 41. Dieminger, W., Pindlay, J.W., Jordan, E.G., Kirby, S.S, and Judson, E.B., Martyn, D.P., Monro, G.H., Proc. Phy. Soc. B., 64, 142 (1951) The8is i n University of Cambridge (Quoted from Whale (1950) ) (1950) (1950) (1955) Loc. c i t . Proc. I.R.E., 2£, 753 Proc. Roy. Soc. A., 201. 216 (1950) Proc. Roy. Soc. A., 202. 208 (1950) Section VII Ghosh, S.P., Thesis Calcutta University (Quoted from Mitra (1948) ) (1958) Section VIII Niblock, P.A., Loc. c i t . (1951) Section IX Appleton, E.V. and Naismith, R., Proc. Roy. Soc. A., I50, 685 (1955) Best, J.P., Farmer, F.T., and Ratcliffe, J.A., Chapman, S., Tremellen, K. and Cox, J., Whale, H.A., Loc. c i t . (1958) Proc. Phy. Soc., 4£, 485 ( l 051b) Jour. Instn. Elect. Engrs., 9jt, 200 (19^ 7) Loc. c i t . (1950) Section X Appleton, E.V. and Weekes, K., Proc. Roy. Soc. A., 17_1, 171 (1959) Budden, K.G., Ratcliffe, J.A., and Wilkes, M.V., Chapman, S., Chapman, S., Pierce, J.A., Proc. Roy. Soc. A., 17JL, 188 (1959) Proc. Phy. Soc. 4£, 26 (1951 A) Loc. c i t . (I951b) Phy. Rev., J l , 698 (1947) 42. BIBLIOGRAPHY PERIODICALS Appleton, E.V., "Some Notes In Wireless Methods of Investigating the Electrical Structure of the Upper Atmosphere," Proc. Phy. Soc., 42, 521 (1950). Appleton, E.V., "Regularties and Irregularities i n the Ionosphere," Proo. Roy. Soo. A., 162, 451 (1937). Appleton, E.V., and Beynon, W.J.G., "The Application of Atmospheric Data to Radio Communication Problems! Part I," Proc. Phy. Soc., 52, 5I8 (1940) Appleton, E.V., and Naismith, R., "Some further Measurements of Upper Atmos-pheric Ionization," Proc. Roy. Soc. A . , 150, 685 (1935). Appleton, E.V. and Naismith, R., "Scattering of Waves in Polar Regions," Nature, l4_5_, 243 (1939). Appleton, E.V. and Naismith, R., "Normal and Abnormal Region-E Ionization," Proe. Phy. Soc., 52, 402 (1940). Appleton, E.V. and Naismith, R., "The Radio Detection of Meteor Trails and Allied Phenomena," Proc. Phy. S o c , 52, 461 (1947). Appleton, E.V., Naismith, R., and Ingram, L . J . , "The Critical-Frequency Method of Measuring Upper Atmospheric Ionization," Proc. Phy. Soc., ^ 1 , 81 (1959) Appleton, E.V. and Weekes, K., "Lunar Tides in E," Proc. Roy. Soo. A . , 171. 171 (1959). Benner, H . , Grace, C.H., and Kelso, J . M . , "Polarization of Low-Frequency Radio Waves Reflected from Ionosphere," Proc. I.R.E., 38, 951 (1950). Best, J . E . , Farmer, F . T . , and Rateliffe, J . A . , "Studies of Region-E of the Ionosphere," Proc. Roy. Soc. A., 164, 96 (I958). Best, J . E . , Rateliffe, J . A . , and Wilkes, M.V., "Experimental Investigations of Very Long Waves Reflected from the Ionosphere," Proc. Roy. Soc. A», I56, 614 (1956). Booker, H.G., Conference on Ionospheric Physics, Penn. State C o l l . , July 1950 ( Quoted from Briggs (1951) ) . Booker, H.G., Rateliffe, J . A . , and Shinn, D.H., "Diffraction from an Irregular Screen with Applications to Ionospheric Problems," Phi l . Trans. Roy. Soc. A., 242, 579 (1950). Booker, H.G. and Seaton, S.L., "Relation Between Actual and Virtual Iono-spheric Height," Phy. Rev., £7_» 87 (1940). Bracewell, R.N., Budden, K.G., Rateliffe, J . A . , Straker, T.W., and Weekes, K., "The Ionospheric Propagation of Low- and Very-Low Frequency Waves Over Distances less then 1000 K.M.," Proc. Instn. Elect. Engrs., 08, 15 (1951) 4 5 . Breit, G„ and Tuve, M0. "A Test of the Existence of the Conducting Layer," Phy. Rev., 2 8 , 554 (1926) . Briggs, B.H., "An Investigation of Certain Properties of the Ionosphere by Means of Rapid Frequency-Change Experiment," Proc. Phy. Soc. B., 64 255 ( W ) . Briggs, B.H. and Phillips, G.J., "A Study of the Horizontal Irregularities of the Ionosphere," Pro. Phy. Soc. B., 65_, 907 (1950) . Briggs, B.H., Phillips, G.J., and Shinn, D.H., "The Analysis of Observations on Spaced Receivers of the Faiding. of Radio Signals," Proc. Phy. Soc. B., 63_, 106 ( 1 9 5 0 ) . Budden, K.G., Rateliffe, J.A., and Wilkes, M.V., "Very long Waves Reflected from the Ionosphere," Proc. Roy. Soc. A., 171. 188 ( 1959) . Chapman, S., "The Absorption and Dissociative or Ionizing Effect of Mono-chromatic Radiation in an Atmosphere on.a Rotating Earth," Proc. Phy. Soc., 4?., 26 and 485 (1951) . Chapman, S., "The Atmospheric Height Distribution of Band Absorbed Solar Radiation," Proc. Phy. Soc., 5 1 , 95 (1959) . Colwell, R.O. and Friend, A.W., "The D-Region of the Ionosphere," Nature, 157. 782 (1956) . Eckersley, T.L., "Irregular Ionic Clouds in the E-Region of the Ionosphere," Nature, l 4 o , 846 (1957) . Findlay, J.W., "Measurements of Changes in Phase-Paths of Radio Signals in the Ionosphere," Nature, 1 5 £ , 58 (1947) . Findlay, J.W., Thesis in University of Cambridge (1950) ( Quoted from Whale (1950) ) . Ghosh, S.P., M.Sc. Thesis Calcutta University (1958) ( Quoted from Mitra (1948) ) . Gledhill, J.A. and Szendrei, M.E., "Theory of the Production of an Ionized Layer in a Non-Isothermal Atmosphere," Proc. Phy. See. Bv, 63_, 427 ( 1 9 5 0 ) . Halliday, E.C., "The Accurate Determination of Ionospheric Equivalent Heights," Proc. Phy. Soc, 48» 421 (1956) . Helllvell, R.A., "Ionospheric Virtual Height Measurements at 100 Kilocycles," Proc. I.R.E., SL, 887 ( 1 9 4 9 ) . Kirby, S.S. and Judson, E.B., "Recent Studies of the Ionosphere," Proc. L.R.E., 2?_, 755 (1955) . Lovell, A.C.B., "Meteoric Ionization and Atmospheric Abnormalities," Rep. Prog, in Phy., xi , 415 (1947) . Manning, L.A., "The Determination of Ionospheric Electron Distribution," Proc. I.R.E., 5 5 , 1203 (1947) . 4 4 . Millington, G., "Deviation at Vertical Incidence in the Ionosphere," Nature, 16J, 215 (19^9). Mimno, H.R., "The Physics of the Ionosphere," Rev. Mod. Phy., £, 1 ( 1957) . Monro, G.H., "Short Changes in the P-Region of the Ionosphere," Nature, 162. 886 (1948) . Monro, G.H., "Short Period Variations in the Ionosphere," Nature, 165. 812, (19^9). Monro, G.H., "Travelling Disturbances in the Ionosphere," Proc. Roy. Soc. A . , 2 0 2 . 208 ( 1950) . Martyn, D.F., "Cellular Atmospheric Waves in the Ionosphere and Troposphere," Proc. Roy. Soc. A . , 201. 216 (1950) . Niblock, P.A., "Lunar Tides in the E-Region," M.A.Sc. Thesis in University of British Columbia (1951) . Pierce, J . A . , "The True Height of an Ionospheric Layer," Phy. Rev., JJL, 698 (19^7). Pierce, J.A. and Mimno, H.R., "The Reception of Radio Echoes from Distant Irregularities in the Ionosphere," Phy. Rev., 5J_» 95 (1940) . Ratcliffe, J . A . , "Diffraction from the Ionosphere and Faiding of Radio Waves," Nature, l 6 2 , : 9 (1948) . Ross, W. and Bramley, E.N., "Tilts in Ionosphere," Nature, 164, 555 (1949) . Shaw, I . J . , "Some Further Investigations of Ionospheric Gross-Modulation," Proc. Phy. Soc. B., 64, 1 (1951). Sil l i toe, S., "Reflections from the Ionosphere," Canadian Jour. Res,, 11 , 165 (1954) . Tremellen, K. and Cox, J . , "The Influence of Wave-Propagation on the Planing of Short-Wave Communication," Jour. Instn. Elect. Engrs., ^ 4 . 200 (1947) , Wells, H.W., Watts, J . M . , and George, D.E., "Detection of Rapidly Moving Ionospheric Clouds," Phy. Rev., 6£, 54o (1946) . Whale, H.A., "The Structure of Region-E of the Ionosphere," Dissertation submitted for fellowship Trinity College ( 1950) . Wilkes, M.V., "Theoretical Ionization Curves for E-Region," Proc. Phy. Soc., a. 159 (1959) . Wilkes, M.V., "The Theory of Reflexion of Very Long Wireless Waves from the Ionosphere," Proc. Phy. Soc. A. , 175. 145 ( 194o) . B O O K S Fleming, J . A . , "Terrestrial Magnetism and Electricity," Dover Publications, Hew York, I 9 A 9 . "Ionospheric Radio Propagation," National Bureau of Standards Circular 462, U.S. Govt. Printing Offiee, Washington, D.C., 1948. Jordan, E.O., "Electromagnetic Waves and Radiating Systems," Prentice-Hall Inc., New York, I950. Mitra, S.K., "The Upper Atmosphere," Royal Asiatic Society of Bengal, Calcutta, 1948. Terman, F . E . , "Radio Engineering," McGraw-Hill Book Oo., Inc., New York, 1947. (a) Refraction of ray in the ionosphere Parabolic gradient of electron density in the region of maximum ionization. Abnormal region-E ia shown as a thin sharp layer. i •—figure 7-PS.T. 0500 ofeoo o 3<x> I Z Q O ) 5 o o / S O O Occurence of high amooth Eg region with low penetration frequency (i.e. below 4 Mc/s) a. The 250 Kilometer Scale PLATE II S p l i t F-echo. No photographs of a s p l i t i n E-echo were av a i l a b l e . Normal I~echo, about 125 ^ m 8 « PLATE III Strong E-scho about 139 kms., and weak echo about f>2 kms. 1 b. Strong E-echo about 130 kms. and weak echoes about 4g and 58 kms. Strong E-echo about 126 kms. and weak scatter I50, 1S52, and 212 kms. PLATS IV a. Echo about 2>5 Kilometers. Shows very bad fl u c t u a t i o n of envelope shape. Frequency Is .^75 mc/a. b. Front view of transmitter and receiver showing traces on C.R.T. display. I o n o s p h e r i c S t u d i e s H u t , showing 85 f t , p o l e a n d d e l t a a n t e n n a s . 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items