CHARACTERIZATION OF IONOSPHERE WAVEGUIDE PROPAGATION BY MONITORING HAARP HF TRANSMISSIONS IN ANTARCTICA

Transcription

1 AFRL-AFOSR-UK-TR-15-4 CHARACTERIZATION OF IONOSPHERE WAVEGUIDE PROPAGATION BY MONITORING HAARP HF TRANSMISSIONS IN ANTARCTICA *Yuri M. Yamolki SCIENCE AND TECHNOLOGY CENTER IN UKRAINE METALISTIV 7A, KYIV, UKRAINE *INSTITUTE OF RADIO ASTRONOMY NATIONAL ACADEMY OF SCIENCES OF UKRAINE 4, CHERVONOPRAPORNA SR, KHARKOV 61 UKRAINE EOARD STCU P-54/STCU 1-8 Reort Date: Aril 15 Final Reort from 1 January 1 to 31 December 14 Ditribution Statement A: Aroved for ublic releae ditribution i unlimited. Air Force Reearch Laboratory Air Force Office of Scientific Reearch Euroean Office of Aeroace Reearch and Develoment Unit 4515, APO AE

2 REPORT DOCUMENTATION PAGE Form Aroved OMB No Public reorting burden for thi collection of information i etimated to average 1 hour er reone, including the time for reviewing intruction, earching exiting data ource, gathering and maintaining the data needed, and comleting and reviewing the collection of information. Send comment regarding thi burden etimate or any other aect of thi collection of information, including uggetion for reducing the burden, to Deartment of Defene, Wahington Headquarter Service, Directorate for Information Oeration and Reort (74-188), 115 Jefferon Davi Highway, Suite 14, Arlington, VA -43. Reondent hould be aware that notwithtanding any other roviion of law, no eron hall be ubject to any enalty for failing to comly with a collection of information if it doe not dilay a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY) 3. DATES COVERED (From To) 17 Aril TITLE AND SUBTITLE. REPORT TYPE Final Reort 1 January 1 31 December 14 5a. CONTRACT NUMBER CHARACTERIZATION OF IONOSPHERE WAVEGUIDE PROPAGATION BY MONITORING HAARP HF TRANSMISSIONS IN ANTARCTICA 6. AUTHOR(S) *Yuri M. Yamolki STCU P-54 5b. GRANT NUMBER STCU 1-8 5c. PROGRAM ELEMENT NUMBER 611F 5d. PROJECT NUMBER 5d. TASK NUMBER 5e. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) SCIENCE AND TECHNOLOGY CENTER IN UKRAINE METALISTIV 7A, KYIV, UKRAINE *INSTITUTE OF RADIO ASTRONOMY NATIONAL ACADEMY OF SCIENCES OF UKRAINE 4, CHERVONOPRAPORNA SR, KHARKOV 61 UKRAINE 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) EOARD Unit 4515 APO AE PERFORMING ORGANIZATION REPORT NUMBER N/A 1. SPONSOR/MONITOR S ACRONYM(S) AFRL/AFOSR/IOE (EOARD) 11. SPONSOR/MONITOR S REPORT NUMBER(S) AFRL-AFOSR-UK-TR DISTRIBUTION/AVAILABILITY STATEMENT Ditribution A: Aroved for ublic releae; ditribution i unlimited. 13. SUPPLEMENTARY NOTES 14. ABSTRACT The Project wa aimed at exerimentally invetigating the oibility of exciting the ionoheric interlayer duct channel uing owerful radiation from the heater HAARP (Alaka, USA) and EISCAT (Tromø, Norway), a well a from HF broadcating tation RWM (Mocow, Ruia) and CHU (Ottawa, Canada). Major attention wa aid to analying the oibility of exciting the interlayer ionoheric waveguide which uort uer-long range HF roagation with a mall amount of attenuation. To monitor the radiation, a comact-ie receiving comlex wa develoed which i caable of meauring the ignal intenity and ectral characteritic in an off-line automatic mode. Two facilitie have been contructed in the coure of the Project. One wa deloyed in Ukraine at the Low-frequency Obervatory of the IRA NASU (Martova village, Kharkov region) in 1, while another wa intalled at the Ukrainian Antarctic tation Akademik Vernadky in the Antarctic in 13. In all, about 1 hour were ent oberving the radiation from the heater (rimarily EISCAT) and more than 3 hour monitoring ignal from broadcat radio. In a number of cae the ignal trengthening wa detected for the uer-long range radio link (Alaka-Antarctica and Northern Scandinavia-Antarctica) which effect can be regarded a a reult of the waveguide roagation. A ioneering feature of the develoed theoretical model i accounting for the regular ionoheric refraction. The aect-enitive contour in the ionohere and on the Earth urface have been calculated for all the tranmitting and receiving ite for the current ionoheric condition. A oftware ackage wa develoed for the remote control of the receiving comlexe and viual rereentation of the meaurement reult in real-time over the internet. The mot roductive exeriment were erformed during the BRIOCHE heating camaign in June 14. The detailed decrition of thi camaign i included in Chater 4 of the Final Reort. The reult obtained during the Project were ublihed in four cientific aer and reorted at everal international meeting in the USA, Puerto-Rico and Ukraine. 15. SUBJECT TERMS EOARD, Material, mictrotructural characteriation, high temerature 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT a. REPORT UNCLAS b. ABSTRACT UNCLAS c. THIS PAGE UNCLAS SAR 18, NUMBER OF PAGES 33 19a. NAME OF RESPONSIBLE PERSON homa R. Caudill 19b. TELEPHONE NUMBER (Include area code) +44 () Standard Form 98 (Rev. 8/98) Precribed by ANSI Std. Z39-18

3 Science and Technology Center in Ukraine Partner Project 54 CHARACTERIZATION OF IONOSPHERE WAVEGUIDE PROPAGATION BY MONITORING HAARP HF TRANSMISSIONS IN ANTARCTICA Final reort (Full Form) Director, Intitute of Radio Atronomy, National Academy of Science of Ukraine Profeor Leonid Lytvynenko Manager, Project P-54 Profeor Yuri Yamolki Kharkiv-14

4 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 1 CHARACTERIZATION OF IONOSPHERE WAVEGUIDE PROPAGATION BY MONITORING HAARP HF TRANSMISSIONS IN ANTARCTICA Project manager: Yuri M. Yamolki 1. Introduction The nonmonotonic run of the height rofile of the ionoheric electron denity aociated with the multilayered tructure of the ionohere create recondition for formation of interlayer duct channel. Thee natural tructure are caable of uorting the radio wave energy tranfer to long ditance (thouand and ten of thouand of kilometer) with low loe. The valley between the E- and F-layer maxima eem to be energetically otimal from thi oint of view. The characteritic boundarie of uch a duct vary between 1 and 15 km on the bottom and from to 5 km on the to, deending on the wave frequency and geohyical condition (geograhic coordinate, unlit condition, eaon, geomagnetic activity etc.). At HF wave the duct rereent a multimode waveguiding tructure ince the wavelength λ i much horter than the waveguide ie h. The ionoheric duct i characteried by a mall amount of loe owing to the comaratively low colliion frequency of electron ν е whoe value i two or three order of magnitude lower than at height of the main aborbing D-region of the ionohere. Thi i a reaon why the attenuation rate of HF ignal in the interlayer duct i doen of time maller than for ordinary multi-ho roagation ath. The major roblem of uing the ionoheric duct in ractice i the difficulty of exciting the channel from the Earth urface and extracting energy for a ground-baed conumer. The reaon i that the E-layer lay the role of a barrier which hade the above layer from the HF emiion radiated by a ground-baed tranmitter at frequencie below the critical frequency of the E-region. The ignal inut in and extraction from the interlayer duct channel i oible only in the reence of coniderable horiontal gradient of the electron denity or intene cattering irregularitie. For the known geometry of the roagation ath it i oible to elect the mot favorable condition for feeding-in (feeding-out) the waveguide roceeding from the variable unlit condition and hence, the redictable horiontal gradient of the electron denity ditribution. Such condition can be mainly realied for rather long-range radio ath of meridional orientation cloe to the unrie or unet at ionoheric height. Pointing the tranmitting and receiving antenna in certain oblique direction and changing the radiation frequency it i oible to arrange a cenario where the bottom wall of the waveguide (E-region) would exit along the entire radio ath, excet the acending (decending) ection of the trajectory in a given vicinity of the tranmitting (receiving) ite. Such ituation are well known to occur in ractice of radio broadcat and communication when the olar terminator ae almot imultaneouly the tranmitter and receiver ite. Thi can be accomanied by a coniderable increae in the received ignal level. The effect of the energy inut and outut owing to radio wave cattering by intene natural ionoheric irregularitie are of unredictable character. Their aearance i determined by oradic factor and deend, ecifically, on the enhanced turbuliation of the ionoheric lama roduced by variou kind of geohyical erturbation. Eecially frequent uch effect are at high latitude which are to a greater extent ubjected to the influence of magnetic and corucular erturbation timulated by olar wind flux variation. It i evident that redictable inut (outut) of the ueful ignal into (from) the duct uing the effect of cattering by natural tructure of the kind i imoible. In addition, the life time of the neceary irregularitie i unredictable a well and varied, a a rule, from a few econd to doen of minute. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

5 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE Extra chance for exciting the interlayer duct can be rovided by active exeriment on ionohere modification which allow controlling the ace and time tructure of the ionoheric turbulence. Primarily thi concern the ionoheric heating by the owerful HF emiion from a ground-baed tranmitter. A i known, the major reult of the nonlinear interaction between the incident wave and ionoheric lama i generation of the artificial ionoheric turbulence (AIT), whoe aceand-time ectrum can be controlled uing ecial heating regime. A a rule, the horiontal cale of the AIT-containing region i determined by the antenna attern width of the heater and can be etimated to lie between 1 and km, deending on the modified region altitude and the heating frequency. It i evident that the timulated irregularitie are of a tochatic character, however a number of their morhological feature are a riori known. Thee are the location; time of aearance, exitence and relaxation; atial aniotroy due to the magnetied lama; external cale-ie of the turbulence; and even atial ectrum in the cae of ecial heating mode. With allowance for uch determinitic feature, the AIT can be ued for exciting the interlayer channel. The outut of the ignal from the waveguide can be rovided by horiontal gradient roduced by the olar terminator near the receive ite of a uer long-range ath. A ioneering feature of the develoed theoretical model i the account of the regular ionoheric refraction. The aect-enitive contour in the ionohere and on the Earth urface have been calculated for all the tranmitting and receiving ite for the current ionoheric condition. During the meaurement camaign with the ue of radiation from the owerful heater two other effect have been occaionally revealed, which are the combination (Brillouin) ignal cattering by the artificially timulated lama turbulence and excitation of the econd harmonic of the owerful radiation in the HF-modified ionohere. A lanned, a oftware ackage ha been develoed for the remote control of the receiving comlexe and viual rereentation of the meaurement reult in real-time through the internet. Taking into account that the mot owerful heater HAARP and EISCAT are located cloe to the northern olar region (Alaka and Northern Scandinavia, reectively), the mot romiing oition to oberve the waveguide roagation effect of HF ignal to uer-long ditance eem to be Antarctic. An additional reaon in favor of thi region i the urity of the electromagnetic climate becaue of the abence of intene man-caued interference and local thundertorm activity. The firt ucceful exeriment on regitration of ignal from a owerful heating facility in Antarctic region were erformed by the Intitute of Radio Atronomy in. The reult uggeted that the waveguide roagation can aarently be realied Zaliovki et al, [9]. The robe ignal wa the roer emiion from the EISCAT heater which wa cattered by AIT inide the interlayer duct. The receiving ite wa deloyed in Antarctica at the Ukrainian bae Akademik Vernadky. In a number of cae, along with the uual quite table comonent aociated with multi-ho roagation through the idelobe of the heater antenna, rather trong ignal were reliably detected whoe ectra allowed uggeting that they were cattered by AIT roduced by the heater emiion itelf. Mot frequently the cattered comonent wa oberved cloe to the time of olar terminator aage through the radio ath. The effect ha been called the elf-cattering of owerful HF ignal by the ionoheric turbulence timulated by the ame ignal Zaliovki et al, [9]. However thee were trial eiodic exeriment, which have not allowed invetigating in detail variou mechanim of radio ignal roagation. The reent Project wa aimed at comrehenive tudying the uer long-range roagation mechanim through monitoring the HAARP and EISCAT heater emiion with the ue of a ecially contructed automated internet-controllable HF receiver. Poible receiving ite hould been deloyed in Antarctica at the Ukrainian Antarctic tation (UAS) Akademik Vernadky (65 15 S, W). The ditance between HARRP and EISCAT heater and Antarctic oition are 15 6 km and 16 1 km correondently. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

6 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 3. Mobile HF receiving ytem For the Stage 1 of the Project the mobile HF receiving comlex wa created and intalled at the Low Frequency Obervatory of IRA NASU(LFO) for tet and continuou meaurement. The receiving facility ha been budett around the WR-G313i digital receiver of the WiNRADiO communication (Autralia, which ha hown good erformance. The receiver hould be deigned a a PCI card to be inerted into and extender lot of the PC mainboard. A general view of the receiver i hown in Fig..1. Fig..1. General view of the WR-G313i receiver The baic erformance ecification of the receiver are lited in Table.1. Table.1 Performance ecification of the WR-G313i receiver Receiver tye Suerheterodyne receiver with a ignal roceor Frequency range 9 kh to- 3 MH Frequency adjutment accuracy 1 H Dynamic range 95 db (11 db with an inut attenuator) Senitivity.5 µv Frequency band Variable between 1 and 15 H with 1 H te Intermediate frequencie IF1: 45 MH IF: 16 kh (adjutable between 1 and kh) IF1 banda filter 15 kh crytal filter Antenna inut 5 Ω Form factor /3 PCI card, PCI. comatible Weight 33 g Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

7 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 4 To rovide for the required frequency intability of the receiver ΔF/F no wore than 1-8, a temerature-comenated comact-ie crytal ocillator develoed at the IRA NASU will be ued. The ocillator i deigned a a tandard 5 PC device to be inerted in a comuter (Fig..). Fig... Comact-ie temerature-comenated crytal ocillator An active loo antenna develoed at the IRA NASU i to be ued a to receive the ignal. The antenna rovide for a high enitivity within the frequency range 3 to 15 MH. The deign of the loo antenna with the re-amlifier i hown in Fig..3 Fig..3. Active loo antenna (1,) broadband re-amlifier (), and fixing art (3) A ecial oftware ackage develoed at the IRA NASU allow remotely controlling the receiver and real-time data reading through the Internet. To detect the ueful ignal, Doler ectra of the received emiion are comuted within a narrow frequency range (about 1 H) with a ectral reolution ~.1 H. Fig..4 how the uer window interface intended for frequency tuning of the Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

8 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 5 receiver through the Internet and viual rereentation of the current meaurement data in the form of the ignal enveloe (a) and intantaneou (b) and dynamic (c) ectra. Fig..4. Uer window interface of the receiver control and data reading A general view of the receiver with a control comuter i hown in Fig..5. Fig..5. General view of the receiving comlex: 1 - WR-G313i receiver, - reference ocillator, and 3 antenna cable The imilar receiving comlex wa created for intallation in Antarctica on the Stage of the Project. On March 13 thi ytem deloyed at the Ukrainian Antarctic tation Akademik Vernadky and oerate continuouly regitering now emiion from HF broadcating tation and ignal from the HAARP and EISCAT heater. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

9 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 6 3. HF wave cattering by field-aligned lama irregularitie conidering refraction in the ionohere The effect of ionoheric refraction on the cattering of high frequency (HF) ignal by random field-aligned irregularitie in the uer ionohere wa analyed in a frame of a theoretical art of the Project. The oibility of excitation of the ionoheric interlayer waveguide by the aectenitive cattered HF ignal i analyed in detail for the ecific condition of the HF heating exeriment. Aect-enitive cattering of high-frequency (HF) (3-3 MH) and very highfrequency (VHF) (3-3 MH) wave by field-aligned irregularitie of the ionoheric lama are widely ued for diagnotic of turbulent rocee in the near-earth lama of both natural [Lyon, 1965; Bourdillon et al.,1995; Berodny et al., 1997] and artificial [Djuth et al., 6; Hyell, 8; Yamolki et al., 1997; Kolokov et al., ] origin. Theoretical remie for thi method of diagnoing the inhomogeneou tructure of the ionohere were develoed within the ingle-cattering aroximation [Rytov et al., 1987]. A a rule, in the relevant tudie the condition cr H i aumed, where i the cyclic frequency of the ounding ignal, cr i the critical frequency of the ionoheric layer (i.e., the maximum lama frequency max ), and H the electron gyrofrequency. Thi aumtion make it oible to ignore the effect of the regular ionoheric refraction and background geomagnetic field. Conequently, the wave trajectorie are aroximated by traight line, and ionoheric lama i aumed iotroic. However, to be recie, in order to ignore the effect of refraction tronger condition have to be atified, which can be derived a follow. The equation for the incident ray trajectory in a lane-tratified ionohere can be written a [Kravtov and Orlov, 199]: d x( ) in, co ( ) / where i the angle of incidence at the lower boundary of the ionoheric layer,, counted from the vertical. Obviouly, in order for a wave to enetrate through the ionohere (with no reflection oint at which the ray curvature would be the greatet), the following condition mut be atified: cr / co ; while in order to aroximate the ray trajectorie by traight line, an even tronger condition hould hold: cr / co. Here i an examle from ractice. During mot of the exeriment on HF ignal cattering from artificial ionoheric turbulence (AIT) roduced by the Sura HF heater (Nihny Novgorod, Ruia) [Yamolki et al., 1997; Myanikov et al., 1; Kolokov et al., ], the geometry wa uch that 7 (for the robe ignal) and f cr cr /( ) 4.5 MH. Therefore, in order to be able to ignore the effect of refraction, the robe ignal frequency hould be much higher than ~13 MH, which wa rarely the cae. Therefore, for correct analyi and interretation of exeriment of the kind, the refraction effect need be taken into account. It i eecially imortant to account for the ionoheric refraction in analying the cattering of owerful HF emiion on the ionoheric irregularitie roduced by the ame radiation. The fact that HF um wave catter on ionoheric irregularitie i rather trivial and ha been oberved a long time ago (Belikovich, et al.,1975; Erukhimov et al., 198). It wa not clear, however, whether the catter ignal can travel over the large ditance away from the heater, for examle, through the ionoheric waveguide. Such effect wa aarently firt exerimentally dicovered by Zaliovky et al. [9] uing the EISCAT HF heater (Tromo, Norway) and wa given the name elf-cattering effect. In their exeriment the EISCAT tranmiion wa monitored at three greatly diered receiving ite: at the Ukrainian Antarctic Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

10 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 7 tation Akademik Vernadky (UAS); at the Radio Atronomical Obervatory of the IRANASU (RAO) near Kharkov, Ukraine; and near Sankt-Peterburg, Ruia (SPB). Tyically, the received ignal contained two characteritic ectral comonent. One comonent wa a relatively narrow line with inignificant Doler hift variation characteritic of the ky-wave HF roagation in middle latitude. The ectrum of the other comonent wa much broader reminicent of that of a ignal cattered at frequencie above the maximum uable frequency (MUF). The broadband (cattered) ectral comonent exhibited trong frequency fluctuation with amlitude rate occaionally exceeding 1 H. However, the mot notable obervational fact wa that the Doler frequency hift and intenitie of thi comonent varied ractically ynchronouly at all three receive ite with nearly the ame ocillation rate. The author concluded that the oberved Doler hift of the broadband ectral comonent were likely induced within a wave trajectory egment that wa common to all the roagation ath, i.e., where the HF um wave travel from the heater to the cattering region with the AIT. Simultaneou obervation by the EISCAT incoherent catter radar howing trong variation in the electron denity above the HF heater are uorting the uggeted mechanim. However, the exact mechanim of the uer long range roagation of the HF heater ignal (the EISCAT-UAS roagation ditance wa ~16,3 km) till ha not been determined, and it i quite oible that ionoheric refraction may lay an imortant role ince the exeriment at the EISCAT location were conducted at frequencie below the critical frequency of the ionohere ( cr ). Galuhko et al. [8] had alo oberved the elf-cattering effect by monitoring the HAARP tranmiion at everal remote ite in the USA, Euroe, and Arctic uing digital Doler receiver of the Intitute of Radio Atronomy (Kharkov, Ukraine) and Lowell Digionde. The aim of the current Project invetigation i to develo a general theory for the aectenitive cattering of electromagnetic wave taking into account ionoheric refraction effect on the incident and cattered wave trajectorie. We will ecifically analye a oibility of ignal channeling in a given direction a a reult of uch cattering. The work on HF ignal traing in ionoheric waveguide due to cattering wa ioneered in 197 (e.g., Erukhimov et al., 1975 Gurevich et al., 1975). Our aroach, however, i quite different from thoe tudie a we will analye the range of incidence angle at the lower boundary of the ionohere reonible for channeling ignal in a given direction rather than evaluate a traing coefficient for HF wave. The gyrotroic effect of the ionoheric lama on the radio wave roagation will be aumed negligibly mall. Note that thi i quite legitimate firt-te aroach ued in earlier work a well (e.g., Gurevich et al., 1975). Let a lane monochromatic electromagnetic wave be roagating in a horiontally tratified ionohere from the lower half-ace through the ionoheric layer (ee Figure 3.1). The k () k (), k (), k (). Here wave i characteried by a frequency and wave vector ( ) k i ( ) x ( ) k in co, k i ( ) y ( ) k in in and k i ( ) k co, where k / c, and angle and determine the direction of the wave vector at the lower boundary of the ionohere. The wave frequency i aumed to be much higher than the electron gyrofrequency H, i.e. H. Thi make it oible to treat the ionohere a an iotroic medium, and to aume the roagating wave to be tranveral. In order to avoid any confuion, the following nomenclature will be ued in our analyi. The HF wave will be regarded a incident and cattered with reect to the cattering on the ionoheric irregularitie. With reect to the roagation in the ionohere, the wave will be conidered a direct and reflected. x y Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

11 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 8 Fig Coordinate ytem for the roblem. X -axi oint toward the geomagnetic north ole. h (i) vector indicate the direction of the local magnetic field. k () i the incident wave vector at the lower ionoheric boundary. The ionohere will be ecified a a lane-tratified colliionle dielectric medium containing random irregularitie: ( ) ( r, ), ; ( r, ) 1,. Here r x, y are horiontal coordinate; ( ) i a regular (i.e., without irregularitie) comonent of the dielectric ermittivity of the ionohere, ( ) 4e N( ) ( ) 1 1, (1) m where N ( ) i a regular electron denity rofile, and e and m are the electron charge and ma, 4e N( r, ) reectively; and ( r, ) i a random addition due to electron denity fluctuation m N( r, ), characteried by ero mean N( r, ) and variance [ N( r, )] N ( ) (the angular bracket... tand for tatitical averaging). Note that the fluctuation ( r, ) are related to the relative electron denity variation N ( r, ) / N ( ) a: ( r, ) [1 ( )] N( r, ) / N( ). () Electron denity irregularitie N( r, ) (or irregularitie in the dielectric ermittivity ( r, ) ) of the magnetied lama in the uer ionohere are highly aniotroic, tretched along the geomagnetic field direction h. The oition of the unit vector h in the lane of the geomagnetic meridian y (ee Figure 1) i ecified by the inclination angle I ( 9 I 9), counted from the horiontal lane [Akaofu and Chaman, 197]. Poitive value of I correond to h Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

12 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 9 ointing downward (northern hemihere), while the negative one correond to the uward direction (outhern hemihere). Then, for the elected ytem of coordinate (the x -axi in Figure 3. 1 oint to geomagnetic north) we have h h h, h coi,, in I. (3) x, y In order to calculate characteritic of cattering of the incident electromagnetic wave by uch ionoheric irregularitie, let u aly the ingle cattering aroximation method [e.g., Rytov et al., 1987]. In thi cae it i aumed that [ ( r, )] ( ), and therefore, it i oible to ignore the effect of the ionoheric irregularitie on the wave trajectorie and calculate them within the geometrical otic aroximation. Trajectory arameter of the aect-enitive cattered ignal. It i known [Gerhman et al., 1984] that HF wave incident on the field-aligned irregularitie of the uer ionohere are redominantly cattered in the direction determined by the o-called aect condition: K( ) h. (4) Here K ( ) i the cattering vector, which i equal to the difference of the wave vector of the cattered k ( ) and incident k ( ) field at the cattering oint : K( ) k ( ) k ( ). (5) A i evident from Figure 3., equation (4) i atified when the angle made by the magnetic field () vector with the incident and cattered wave vector are equal. A a reult, the et of vector k form a characteritic angular cone around the magnetic field with the aex angle. (i) Fig. 3.. Scattering geometry. Vector k () and k are, reectively, the incident and cattered wave vector at the cattering oint ; h indicate the direction of the local magnetic field. In the coordinate ytem hown in Figure 3.1, and with account of (5), equation (4) can be written a: h (in co in co ) h (co co ), (6) x () () (i) (i) where, and, are the angle ecifying the wave vector oition of the cattered (uercrit ) and incident (uercrit i ) field at the cattering oint (hereafter, ymbol ( ) i omitted to imlify the equation). Solving equation (6) for in and co, it i oible to derive the direction of the wave vector of the cattered field. Uing the relation in co 1, two oible olution are obtained: Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

13 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 1 ( co ) 1, h h x 1 ( co ) h ( h x ( i in ) co h ( i co ) ) h x ( co ) (7а) ( in ) 1, h h x ( h x 1 co ( i co h ) x co h ( h x ( i in ) in ) co h h x ( (co ( i co ) ) ( )co co ), ) 7b) h ( h x ( i co ( Here the to ign tand for ) ) 1 co h in ) h x ( (co ) (, while the lower ign tand for co ). ). It i alo taken into account that according to the equation of geometrical otic for a lane-tratified medium [e.g., Kravtov and Orlov, 199]. Note that the incident wave can be cattered either uward ( ) ( co ), or downward ( co ), i.e., [, ]. Therefore, the olution ace of 1, 1, 1, () equation (7) with reect to i limited by the condition: in. (8) 1, Another neceary condition for the exitence of a real olution of equation (7) i a non-negative argument of the quare root in equation (7), i.e., co The value of in 199] in co co co in. (9) h x ( i ) at h i related to in at through Snell law [Kravtov and Orlov, ( ) in ( ) ( ) in ( ). (1) Whence it follow that if in ( ) at a certain altitude, then the value of in ( i ) i equal to 1 at that altitude (i.e., ( ) / ). Uing equation (1) thi condition can alo be written a: co ( ). (11) Thu, equation (11) determine the critical reflection height cr for the ray with the incidence angle at the lower boundary of the ionoheric layer. Thi i why when we determine co ( ) two ituation are oible, deending on the ratio between co and cr (ee Figure 3.3). If co cr, then the wave enetrate through the ionoheric layer (no reflection). In thi cae, within the entire volume of the ionoheric layer with irregularitie, the cattering occur only at the acending art of the wave trajectory (the direct wave i cattered alone), and the value of co ( ) according to [Kravtov and Orlov (199)] i: co ( ) co ( )/. (1) If co cr, then the wave i reflected at cr, determined by equation (11), and the range of i limited a cr. A a reult, two wave arrive at the cattering oint, one being cattered on the acending art of the trajectory (direct wave), the other on the decending ath (reflected wave). Then for co ( ) one get: co ( ) co ( )/, (13) where + and - tand for the direct and reflected wave, reectively. 1. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

14 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 11 m cr d 1 k d 1 m cr k d d k r r 1 k 1 ( k ( x x a) b) Fig Trajectorie of the incident wave for co cr ( 1 ) and co cr ( ), where cr i the critical frequency of the layer (lama frequency at the height of the layer maximum ). In the firt cae (a), the direct wave ( ( ) ) alone reache the m cattering height, while in the econd cae (b) two wave reach the height, ecifically, the direct wave ( ( ) ) and the wave reflected at cr ( ( ) ). d Now, conider the cattered wave. Let the background electron denity rofile N ( ) be a mooth r d1 function with a ingle maximum N m ( m ) at m (ingle layer model). If cr, then the wave cattered downward ( co ) and uward ( co ) both reach the lower boundary of the ( ) ionoheric layer at with an angle [ /, ], which can be found from an equation imilar to equation (1), vi. in ( ) in ( ). (14) If cr, then only the wave cattered at angle ( ) can reach the lower boundary of ( ) the ionohere ( ). Here cr [, / ] i a certain critical angle whoe value, with account of equation (1) and Snell law (1), can be determined a: cr incr. (15) ( ) If ( ) cr, the cattered wave goe into the uer half-ace and then enetrate through the ionohere. Note that in the cae of a multi-layer ionohere (e.g., E - and F -layer are reent) the wave cattered in certain direction can be traed by the interlayer ionoheric waveguide. Such a oibility will be invetigated in the next ection, but firt let u analye the amlitude characteritic of the aect-enitive cattered ignal. Scattering cro-ection. The effective differential cro-ection of a random medium i ued to characterie the energy cattered by a unit volume into a unit olid angle in a given direction with a unit flux denity of the incident radiation [e.g., Rytov et al., 1987]. In the aroximation of a ingle cattering of electromagnetic wave in random iotroic medium with a regular dielectric ermittivity cont, the cattering cro ection i [Rytov et al., 1987; Gerhman et al., 1984]: cr Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

15 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 1 4 k Q( K( )) P( ) ( K( )). (16) Here ( K ( )) i a three-dimenional atial ectrum of fluctuation ; K ( ) i the ( ) ( ) cattering vector given by equation (5); and ( ) 1 k e P i a olariation factor, ( ) k where e ( ) i the unit vector of the electric field olariation of the incident wave. Uing the ame geometrical otic aumtion a in Rytov et al. [1987], it can be hown that equation (16) i alicable to the analyi of the cattering of a ingle quai-lane wave in the medium with mooth atial variation in ( ). The factor P ( ) i deendent on the incident wave olariation. For intance, according to [Rytov et al., 1987] in the cae of a linear olariation P( ) in ( ), (17) and for a circular olariation 1 P( ) (1 co ( )), (18) where ( ) and ( ) denote the angle made by the cattered field wave vector k ( ) with the olariation vector e ( ) and wave vector k ( ) of the incident field, reectively. The aniotroic ower law model (e.g., Gerhman et al. [1984]) i a conventionally ued aroximation for the atial ectrum (K ) of the fluctuation ( r, ) in the magnetied lama of the uer ionohere for the inertial interval of the wave number. In thi model, K ( )) can be written a ( / ( K L ) ( K L ( K, K ) C ( )1 ), (19) where C ( ) ~ ( ) i a normaliation factor, with ( ) rereenting the variance of the fluctuation ; K and K are longitudinal and tranveral (with reect to h ) comonent of the cattering vector K ; L and L are characteritic longitudinal and tranveral external caleie of the ionoheric turbulence with L L ; and 3 4. If condition (4) i atified, then equation (16) yield 4 k / Q( ) P( ) C ( )1 K ( ) L, () where h x K( ) k in in (in co in co ) h (1) in in co( ). Equation (1) i inalicable in the ecific cae of the horiontal magnetic field, h, (at the magnetic equator). It can be hown that in thi cae K ( ) become K ( ) k in in in in co( ). Uing equation (), C ( ) can be exreed a Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

16 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 13 4 ( ) N ( ) N ( ) C ( ) ~ ( ) (1 ( )). () 4 N ( ) N ( ) Therefore, from equation () it i oible to determine the cattering cro-ection in the direction defined by the aect condition (4) for a lane electromagnetic wave roagating in an ionoheric layer with aniotroic electron denity irregularitie. Note that ince ( ) and ( ), all the term in equation () can be exreed through the oition angle of the incident and cattered field wave vector at the lower boundary of the ionoheric layer. Dicuion. A an examle, let u firt conider cattering of a lane circularly olaried electromagnetic wave roagating vertically in an ionoheric layer with aniotroic irregularitie. The wave frequency i aumed to be lower than the critical frequency, cr. Note that thee are tyical condition for the HF heating of the ionohere. In a ingle layer model ionohere two comonent will be cattered in the height range cr ( cr i determined from (11) with co 1), which are the direct and reflected wave. Then equation (7) with account of equation (13) yield: h h x co co d, r, (3а) h h co x in h xh co d, r, (3b) ( h co ) h x where the + and - ign tand for the cattering comonent of the direct (ubcrit d ) and reflected (ubcrit r ) wave, reectively. A can be een from equation (3b), in the northern hemihere ( h ) the direct wave i h x () cattered outhward ( 9 9 ), while the reflected i cattered northward (9 7). In the outhern hemihere the ituation i revered. Near the magnetic equator ( h ), a follow from equation (7b), both comonent are cattered redominantly ( ) within the lane erendicular to the geomagnetic field line at all angle [, ]. To analye the orientation of the cattered field wave vector at the exit from the ionoheric layer it i convenient to introduce an angle out. Then, according to Snell law and uing equation (3b) and (1), one get in 1 ( ) / co ( ) h xh out. (4) ( h h co ) x A an examle, Figure 4 how ioline for 75(55 ), 45 (5) and (18 ), which are lotted in the coordinate ( ) / and h x () out. The inclination angle I of the co and in I ), which h geomagnetic field i et equal to 77.5 (recall that I correond to the location of the EISCAT HF heater ( 6935 N, 1914 E). In thi cae, the wave vector of the cattered comonent at the lower boundary of the ionohere ( ) can be ignificantly off the vertical deite trongly vertical roagation of the rimary wave. The value of thi offet increae a decreae, (i.e., with decreae of the cattering altitude ), and/or a the cattering direction aroache the lane of the magnetic meridian. In thi examle the maximum offet i equal to I 5, for and,. Analyi of equation (4) how that uch a deendence i valid for the geomagnetic field inclination I / 4. Otherwie, if I / 4, the maximum offet of the aect-enitive cattered wave from the vertical, Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

17 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 14 out / (i.e., the cattering within the horiontal lane) will occur for and co / h. h x ( θout, ( θout, -5 (), a) b) (), d -5 Fig Angular ditribution of the value log[ Q ( ) / Q ( )], calculated uing equation (5) 1 with 11/ 3 for L / 1 (а) and L / 5 (b). The olid, dahed and dotted line correond to the level ( ) /. 3,. 75 and. 99, reectively. Calculation were made for the vertical incidence at the latitude correonding to the EISCAT location (I ). The value of the cattering cro-ection increae with (i.e., with the cattering () height ), reaching it maximum at 1, which correond to. To calculate the cattering cro ection in the direction defined by equation (3) we combine equation (18), (1), and (3) and Snell law (equation (1)) obtaining for the vertically roagating wave h h x co P( ), ( h h co ) K x 4k ( ) h x co ( ). h h x co Subtitution of thee exreion together with equation () into () yield ( ) h h x co Q( ) ~ Q ( ) 4 ( h h co ) 4k h xl ( )co 1, h h x co 4 k N ( ) where Q ( ). For quiet ionoheric condition it i uually aumed that the N ( ) lama denity fluctuation are roortional to the background electron denity, i.e., 4 N ( ) ~ k cont N. Thu one can write Q ( ) ~ N. N ( ) x / out (5) Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

18 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 15 Figure 3.5 how the angular ditribution of the value log[ Q ( ) / Q ( )], calculated with 1 equation (5) with 11/ 3 and L / 1 (to anel), and L / 5 (bottom anel) for the vertical incidence cae for the latitude correonding to the EISCAT location ( I ). Here f / c k i the free ace wavelength. The reult are hown for the cattering of the / reflected wave (cattering in the northward direction, () 9 ). For the direct wave (cattering () in the outhward direction, 18 9 ) the angular ditribution are comletely ymmetric. In order to be able to at leat roughly relate thee data to the cattering height (or the correonding lama frequency ) ) Figure 5 how ioline for ( ) /. 3 (olid ( () () line),. 75 (dahed line), and. 99 (dotted line). Note that out, and ( ) are related through equation (4). A i evident from Figure 5, the value of the cattering cro-ection increae with (i.e., with the cattering height ), reaching it maximum at 1, which correond to. Thi i becaue ( ) increae with height (ince it i aumed that ( N ) ~ N ( out 4 ) ~ ( N ) ), and alo becaue the econd term in the quare bracket of equation (5) get maller (ince ( ) a cr ). The latter effect can be treated a an increae of the ignal wavelength in the ionohere a it aroache the reflection oint ince k ) / ( ) k ( ). Aarently, thi i alo the reaon for the widening of the ( () aimuthal ditribution of the cattering cro-ection with the decreae of out. For examle, already for 51 the cattering cro ection i ractically iotroic in the aimuthal out lane, and maller L / value lead to greater value of the cattering cro-ection. Thi would ugget that in the reence of everal ionoheric layer (e.g., E and F ) a certain fraction of the cattered wave energy can be catured by the interlayer ionoheric waveguide. Since uch waveguide roagation i characteried by a very mall attenuation, it i quite lauible that thi mechanim uorted the uer long ditance roagation of the EISCAT ignal oberved in the elf-cattering exeriment [Zaliovki et al., 9]. The otential of exciting the ionoheric interlayer waveguide by the aect-enitive cattered HF ignal require greater invetigation. It hould be noted that the role of cattering in HF radio wave traing into ionohere waveguide wa rather intenively examined a early a in 197 [e.g., Gurevich et al., 1975; Erukhimov et al., 1975]. However, in contrat to the reviou tudie, we will be rimarily intereted in determining the range of and reonible for channeling cattered ignal in a given direction rather than in etimating the traing coefficient for an incident wave. Let the ionohere be rereented by two layer, e.g., E and F, each characteried by their reective critical frequencie cre and crf and critical height me and mf. We denote the minimum lama frequency inide the E-F valley at the height v a v. Further aume that crf, i.e., the wave with frequency i reflected from the F -layer (the uer wall of the waveguide) at any incidence angle. Obviouly, the interlayer waveguide can be excited by the wave cattered in the height range me mf only. Therefore, it i neceary for the direct wave to enetrate the E -layer, i.e., co cre. (6) A follow from Snell law, in order for the cattering comonent to be confined inide the waveguide, i.e., to be reflected from the E-layer (the lower wall of the waveguide), the following condition mut be met: ) in ( ) ( ) (7a) ( cre Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

19 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 16 which, with account of equation (1), can be recat in a form imilar to equation (15): cre in ( ), (7b) ( ) where in ( ) i determined by equation (7b). Since in ( ) 1, it follow from equation (7b) that the cattering hould occur within the height range me, where, ], F F [ v mf and can be determined from ( F ) cre. Further analyi will be carried out for the ecific condition of the elf-cattering exeriment conducted by Zaliovki et al. [9]. We are rimarily intereted in the oibility of channeling the EISCAT ignal inide the interlayer waveguide in the direction of the receive () ite located at UAS, SPB, and RAO. In the choen coordinate ytem, 135 for the UAS () tation and 5 for the other two tation. In mot heating eion the EISCAT heater tranmitted toward magnetic enith at one or two cloe frequencie around 4.4 MH. The meaurement from the collocated ionoonde howed the reence of trong E layer, with the critical frequencie cloe to and even above the heating frequencie. For the analyi fcre 3.9 MH and f H 1. 4MH are aumed and the calculation are not limited to the vertical incidence cae, making it oible to invetigate the effect of the angle of direct wave incidence. A known [Gurevich, 7], artificial ionoheric turbulence i effectively roduced by the HF heating in the height range with the lower boundary, determined from the uer hybrid reonance condition f f ( UH ) f H ( UH ), where f H i the gyrofrequency; and the uer boundary cr, determined by the wave critical reflection condition equation (11). For thi reaon the lama frequency f ) in the calculation i aumed to vary within the cattering region ( from f ( ) f f ( ) 3. 79MH, to f ( ) f 3. 9MH. Figure 6 how reult UH H UH cr UH cre of the calculation for uch condition in the ytem of coordinate with (counted radially), and ( ) (counted counterclockwie). The lot how the ditribution of in value which atify the condition of excitation of the interlayer ionoheric waveguide (ee equation (7b)). The () calculation were made for f 4. 4MH, 135 and f 3. 79MH (Figure 6а) and 3.87MH (Figure 6c). Figure 6(b) and 6(d) how the reective ditribution of the value f 1log[ ( ) / Q ( Q )] for 11/ 3 and L / 1. The value are given in graycale (ee the ( ) legend on the right). A follow from equation (7) and (1), the ditribution of in for () 5 will be the ame a in Figure 6 (a) and (c), while the ditribution of log[ Q ( ) / Q ( )] will look like mirror image of Figure 6 (b) and Figure 6 (d) with the 1 reflection line at 18. Note that limit of the region from which the cattered ignal are catured by the interlayer waveguide are determined at the level equation (7b)), which i aroximately equal to.754 for f cre in ( ) (ee ( ) f 3.79 MH and.99 for 3.87 MH. The dahed line in Figure 6 how the EISCAT antenna beam at the half-ower level. A can be een, the ignal roagation through the interlayer duct channel in the aimuthal () () direction 135 and 5 i oible for the given orientation of the heater antenna Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

20 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 17 only for a very narrow range of angle near 15 and 18, with characteritic ie (for f 3.87 MH) of 15 in aimuth and le than 1 in (ee Figure 3.6 а). a) b) c) d) ( ) Fig Ditribution of the in value which atify the condition of excitation of the ionoheric interlayer waveguide (equation (7b)). Calculation were made for f 4. 4 MH, () 135 and f MH (а) and f 3. 87MH (c). Panel (b) and (d) how the reective ditribution of the value log[ Q ( ) / Q ( )] for 11/ 3 and L / 1. The value are 1 hown in graycale (ee legend on the right ide) in the coordinate ytem with (hown a radiu) and (counted counterclockwie). The dahed line how the EISCAT antenna beam at the half ower level. It can be een that for the given orientation of the heater antenna, ignal roagation in the interlayer waveguide toward the receive ite i oible only for a very narrow range of angle near 15 and 18, with characteritic ie (for f 3.87 MH) about 15 in aimuth and le than 1 in. Accordingly, the linear ie of the cattering region at km i aroximately 45 to 5 km acro and 3 to 4 km along the magnetic meridian. Note that, a follow from equation (7), thi cae correond to the cattering of the ionoherically reflected ignal. The har cutoff at 15 i determined by condition (6) for the direct wave to enetrate the E -layer. The value 1log[ ( ) / Q ( of Q )] in thi angular range varie inignificantly, from about 4 to 1 (ee Figure 3.6 b). A height cr f increae, i.e., a the cattering height aroache the critical reflection, the angular and linear ie of the cattering region decreae by aroximately a factor of (ee Figure 3 6 c), while the cro-ection change inignificantly (ee Figure 3. 6 d). The above analyi uort the mechanim of the elf-cattering effect that wa uggeted by Zaliovky et al. [9] for the exlanation of their exerimental reult of Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

21 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 18 monitoring the EISCAT tranmiion at three remote location (SPB, RAO, and UAS). Recall that during the exeriment the variation of the Doler frequency hift recorded at all the ite were almot identical. During the exeriment ignificant variation of the electron denity were alo oberved by the EISCAT incoherent catter radar in the height range from 1 to km. Such variation can ignificantly affect the Doler frequency hift and amlitude of the HF heating ignal a wa demontrated by the author uing comuter modeling. With the above analyi we have demontrated that the angular ie of the cattering region reonible for the excitation and ubequent channeling of the HF ignal through the interlayer waveguide to the receive ite are rather mall (ee Figure 3.6). Therefore, it i reaonable to aume that all remotely recorded ignal firt travel through the ame irregularitie in the lower ionohere, and are equally affected by them. Then, the ignal reflected from the ionoheric F-layer are cattered by field-aligned irregularitie roduced by the ame EISCAT heating tranmiion and roagate toward the receive ite through the ionoheric waveguide. Since the cattering region guiding the ignal into the interlayer duct channel for roagation into different direction i the ame, it i lauible to exect that the variation of the Doler frequency hift and ignal amlitude recorded at greatly diered receive ite will be very imilar, a wa oberved during the elf-cattering exeriment reorted by Zaliovki et al. [9]. Concluion. In thi chater the ray otic aroximation and mall erturbation method were ued to analye the effect of refraction on the cattering characteritic of HF ignal cattered by random field-aligned irregularitie of the ionoheric lama. The calculation are carried out for an iotroic lane-tratified (on average) ionohere, i.e., the incident and cattered wave were both aumed to roagate along unerturbed trajectorie. The equation of the o-called aectenitive cattering i derived which relate the trajectory characteritic of the incident and aect-enitive cattered wave. The cattering cro-ection i calculated within the Born aroximation for a lane electromagnetic wave roagating in the ionoheric layer with aniotroic electron denity irregularitie. For the vertical incidence cenario the intenitie of the aect-enitive cattered wave (cro-ection) and it exit angle are calculated a function of the cattering direction and height. A oibility of excitation of the interlayer ionoheric waveguide by the aect-enitive cattered ignal with a elected orientation of the horiontal rojection of the wave vector i invetigated in deendence on the angle of direct wave incidence. It i demontrated that uch oibility can be imlemented uing owerful HF heater, like EISCAT or HAARP. Under certain condition (e.g., ecific orientation of antenna beam), the cattered HF ignal can be traed inide the waveguide and ubequently channeled to very long ditance becaue of the very mall attenuation characteritic of the interlayer waveguide. Thu, the aect-enitive catter mechanim can exlain the elf-cattering effect which wa oberved uing the EISCAT tranmiion by Zaliovky et al. [9]. Therefore, there exit the otential for the ractical alication in HF communication, although further tudie are required for a more reliable aement. In articular, it i neceary to conider the uggeted mechanim with account of the magnetic field effect which can influence eentially roagation of the incident and cattered wave. Addreing thi iue roerly would require alication of full cale 3D numerical ray-tracing. The author lan to carry out the reective analyi in the future and reent the reult in a earate ublication. A a firt aroximation, however, we think that neglecting the geomagnetic field effect on radiowave trajectorie i a legitimate aroach, which wa alo taken by other author [e.g., Gurevich et al., 1975]. The reult of thi invetigation can be ueful for the analyi and interretation of the exeriment on HF ignal cattering from artificial and natural ionoheric irregularitie in the ionohere, for the develoment of new method of diagnotic of the near-earth lama turbulence, and alo for the invetigation of the mechanim of uer-long range roagation of HF electromagnetic emiion. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

22 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 19 Reference to Chater 1-3 Akaofu, S. I., and S. Chaman (197), Solar-Terretrial Phyic, Clarendon Pre, Oxford. Berodny, V. G., P. V. Ponomarenko, and Y. M. Yamolki (1997), Ionoheric cattering of broadcat ignal a a ource of tochatic interference to highly enitive HF radio ytem, 1- th International Zurich Symoium and Technical Exhibition on Electromagnetic Comatibility, February 18-9, Zurich, Switerland, Bourdillon, A., C. Haldoui, and J. Dellone (1995), High-frequency Doler radar obervation of magnetic aect irregularitie in the midlatitude E region ionohere, J. Geohy. Re., 1, , doi:1.19/95ja179. Djuth, F. T., B. W. Reinich, D. F. Kitroer, J. H. Elder, A. Lee Snyder, and G. S. Sale, Imaging HF-Induced irregularitie above HAARP, Geohy. Re. Lett., 33(4), L417, doi:1.19/5gl4536, 6. Erukhimov, L. M., S. N. Matyugin, and V. P. Uryadov (1975), Radio-wave roagation in an ionoheric wave channel, Radiohyic and quantum electronic, 18(9), , doi: 1.17/BF Galuhko, V. G., A. V. Kolokov, V. V. Panukhov, B. W. Reinich, G. S. Sale, Y. M. Yamolki, and A. V. Zaliovky (8), Self-cattering of the HF heater emiion oberved at geograhically diered receiving ite, Antenna and Proagation Magaine, IEEE, 5, Gerhman, B. N., L. M. Erukhimov, V. Yu. Kim, V. P. Uryadov, and E. E. Tedilina (1975), Influence of cattering on radio-wave traing in ionoheric waveduct, Radiohyic and quantum electronic, 18(9), , doi: 1.17/BF Gerhman, B. N., L. M. Erukhimov, and Yu. Ya. Yahin (1984), Wave Procee in Ionohere and Sace Plama, Nauka, Mocow, (in Ruian). Gurevich, A. V. (7), Nonlinear henomena in the ionohere, PHYS-USP, 5, Hyell, D. L. (8), 3 MH radar obervation of artificial E region field-aligned lama irregularitie, Annale Geohyicae, 6, Kolokov, A. V., T. B. Leyer, Yu. M. Yamolki, and V. S. Beley (), HF um-induced large cale radial drift of mall cale magnetic field-aligned denity triation, J. Geohy. Re., 17, 197, doi:1.19/1ja154. Kravtov, Yu. A., and Yu. I. Orlov (199), Geometrical otic of inhomogeneou media, Sringer- Verlag, Berlin; New York. Lyon, G. F. (1965), The aniotroy of ionoheric irregularitie deduced from VHF catter meaurement, J. Atmo. Terr. Phy., 7, Myanikov, E. N., N. V. Muravieva, E. N. Sergeev et al. (1), Satial ectrum of the artificial ionoheric diturbance induced by owerful HF radiowave. Radiohyic and Quantum Electronic, 44, Rytov, S. M., Kravtov, Yu. A., and Tatarkii, V. I., Princile of Statitical Radiohyic, Vol. I- IV, Berlin, Sringer-Verlag, Zaliovki, A. V., S. B. Kahcheyev, Y. M. Yamolki, V. G. Galuhko, V. S. Belyey, B. Iham, M. T. Rietveld, C. La Ho, A. Brekke, N. F. Blagovehchenkaya, and V. A. Kornienko (9), Self-cattering of a owerful HF radio wave on timulated ionoheric turbulence, Radio Science, 44, RS31, doi: 1.19/8RS4111, 1-1. Yamolki, Y. M., V. S. Beley, S. B. Kaheev, A. V. Kolokov, V. G. Somov, D. L. Hyel, B. Iham, and M. C. Kelley (1997), Bitatic HF radar diagnotic of induced field-aligned irregularitie, J. Geohy. Re., 1, Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

23 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 4. Studie of the Ionoheric Turbulence Excited by the Fourth Gyroharmonic at HAARP and ionoheric wave guide excitation There are the reult of exeriment conducted during the HAARP BRIOCHE June 14 camaign, whoe objective wa to tudy the develoment of artificial ionoheric turbulence. During the exeriment, the heating frequency wa teed u near the 4 th gyroharmonic, and the ower of the heating HF radiation wa varied. Our diagnotic included: meaurement of haederived Slant Total Electron Content uing the L1/L GPS ignal received at HAARP; meaurement of Stimulated Electromagnetic Emiion (SEE) conducted 15 km away from the HAARP ite; ionogram from HAARP digionde and reflectance data from Kodiak radar; and detection of the HAARP HF radiation cattered into the ionoheric channel and roagated for 15.6 Mm to the receiver at Antarctica Peninula. Additionally, a new method of diagnotic wa introduced during thi camaign. Thi method wa develoed of the IRA NASU team in a frame of the EOARD STCU Partner Project P-54. HAARP radiated HF wave can be cattered into the ionoheric waveguide by artificial irregularitie, and can roagate along the waveguide like a ound wave in the whiering gallery mode. The waveguide wa oriented along the Earth terminator which wa aing over the tranmitting and receiving ite imultaneouly. Furthermore the electron denity gradient formed in the ionohere when the terminator aage occur near the receiving ite could catter the HF ignal from the waveguide to the groundbaed receiver. We were able to detect the cattering HAARP ignal on the ground at Ukranian Antarctic Station (UAS) (coordinate 65.5 S, 64.5 W) at 15.6 Mm from HAARP. The HF ignal were recorded within frequency band 5 H centered at the carrier frequency of the HAARP umed ignal. The data acquiition ytem collected record of the intenity and the Doler ectra of the ignal with time reolution 1 ec and 5 ec. Thi aer introduce a ucceful attemt to ue the effect of long ditance HF roagation for diagnotic of the artificial turbulence. Finally, all the exeriment were diagnoed by the iononoonde, and by the Kodiak coherent radar located 67 km South Wet from HAARP. Ionoheric condition. The two exeriment dicued were conducted during minute each tarting at about 3 UT, i.e around 7.m. local time. During the firt day (6/6/14) the ionohere wa lightly diturbed (δb~5 nt); a noticeable oradic E-layer exited (f E ~4.5 MH); f O F wa in the frequency range MH while the F eak wa located at h m F =7 km. Conidering the oblique roagation of the heating wave, it frequency wa reflected from the F eak thu, for the majority of the exeriment, the heating wave wa trongly aborbed in the ionohere due to anomalou abortion. An excetion wa the time around 3:4 UT when the ionogram revealed that f O F droed below 5.4 MH, thu the heating frequency exceed the critical frequency, and the ionohere became tranarent to the HF wave. During the econd day (6/7/14) the ionohere wa alo quiet (δb~3 nt) with E layer reent. However, the ionohere wa untable, half of the ionogram reveal that f O F fell to MH, thu the HF frequency wa higher than the critical frequency and abortion wa low and the artificial turbulence wa almot abent, while the ret of the ionogram how f O F >5.7 MH and h m F =7km, and thu at the time the artificial turbulence wa umed. Note that due to the exeriment retriction the minimum heating frequency wa 5.67 MH which wa above the 4 th gyro reonance. In our March 13 exeriment [Najmi et al., 14] we have obtained the reonance frequency 5.76 MH from the SEE ectrum, at which frequency, the DM diaeared. The ionogram how that 5.76 MH wave are reflected at 19 km. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

24 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 1 An etimate baed on the diole model of the geomagnetic field how that in the June 14 R exeriment the 4 th E 19km gyro reonance occur at 5.6 MH, 4 fce 5.76MH R E 5 km, where 5 km i the reflection height of 5.6 MH taken from ionogram. Thu the lowet heating frequency 5.76 MH i cloe to the BUM cutoff which i about 8 kh above the 4 th gyroharmonic, i.e. at 5.68 MH [Leyer et al., 1994]. SEE obervation. Stimulated Electromagnetic Emiion (SEE) ignal were meaured imultaneouly with the STEC uing an HF detector oerated by the Naval Reearch Laboratory 15 km away from the HAARP ite. Figure 4.1 how the ower ectral denity (PSD) of the SEE emiion for each of four choen heating frequencie, 5.67, 5.7, 5.73 and 5.76 MH, which reveal ditinct BUM. At higher heating frequencie BUM diaear. The trace are averaged over a 1 ortion of the heating eriod with contant ERP. The heating frequency i hown by the highet eak in the center at Δ F =, the down hifted maximum (DM) i on the left ide of the heating frequency at Δ F =-1 kh, while the broad uhifted maximum (BUM) i on the right ide of the lot in the range of Δ F =+3-13 kh. In each of the anel of Fig. 4.1, the blue trace correond to the ower.5 kw, green to 5. kw and red to 1 kw er tranmitter (the ower level 5, 5, and 1% reectively). Note that the hae of SEE ignal aturate at the half full ower. Figure 5a,b how the amlitude of the PSD normalied by it eak value a a function of heating ower. Here Fig. 4. a) correond to the BUM amlitude while Fig. 4. b) how the evolution of the DM amlitude. In thoe two cae aturation at 9% level occur at about ½ of the full ower. 3 Fig Power ectral denity (PSD) of the SEE emiion for each of four choen heating frequencie, 5.67, 5.7, 5.73 and 5.76 MH, which reveal ditinct BUM. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

25 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE a) b) Fig. 4.. Behavior of the BUM amlitude. Obervation made by the Kodiak radar. Figure 4.3 a) lot time erie of the SNR of reflected a detected by the Kodiak radar. We choe beam 9 directed toward HAARP heated region, the frequency of the robing ignal i about 1 MH. The ignal reflected by the artificial ionoheric turbulence i centered at 7 km lant range. During the 1 th cycle at 3:1:-3:14:3, the SNR wa trongly reduced ince Kodiak beam 9 wa almot out of the HF heating ot. Thi wa a geometric effect of the angular adjutment of the beam to track PRN 5 and thi final heating cycle wa icked u by beam 1. The HF heating wa witched on at :55 followed by the buildu of the reflected radar ignal over 5 ec. Then a trong SNR of more than 5 db wa detected.the reflection increae by ~1 db around :57 then it di around 3: and 3:4 due to the decreae in the ionoheric lama denity. The econd di wa reflected in the ionogram a dicued above, while the firt di occurred during the time not covered by the ionogram at HAARP. Between 3:11:3 and 3:13:3 the SNR decreae, at 3:14:3 the heater wa witched off and the reflection gradually diaeared. Figure 4.3 b) how time erie of the velocity of the ionoheric irregularitie. Starting at :55:, the irregularitie quickly accelerate to 35-4 m/, eak at ~6 m/ near :57:- :59:, then gradually reduce to ~3 m/. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

26 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 3 a) b) Fig Kodiak radar obervation Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

27 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 4 HAARP ignal detected at UAS. Monitoring of the HAARP HF ignal wa carried out at the Ukrainian Antarctic Station (UAS) Academik Vernadky. Figure 4.4 a) and 4.4 b) how the ectrogram of the detected ignal for June 6 and 7, 14, while figure 4.5 how the intenity of the detected ignal on June 6 th. a) b) Fig Sectrogram a detected in Antarctica at the Akademik Vernadky tation for June 6 (a) and 7 (b). The 8-th (f h = 585 kh) and 9-th (f h = 591 kh) heating cycle, indicated by black arrow at the ectrogram, were contaminated by interference ignal and therefore were not roceed in Fig. 4.6 which reveal the ignal-to-noie ratio of the ignal intenity. Figure 4.5 how that the intenity of detected ignal trongly deend on the heating frequency f h. In fact, for f h lightly Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

28 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 5 above the 4 th gyro the intenity of the detected ignal wa very low, hardly above the noie level. The intenity of the detected ignal went u with f h and eak at 5.79 MH. Fig.4.5. Frequency deendency of the detected ignal intenity. Fig Signal to noie ratio. Further, Fig. 4.6 how how the ignal change with the intenity of the HAARP heater at the different heating frequencie (the blue line). The green line correond to 4 time increae of the heating ower, the linear interolation i hown by the red line in both Fig. 6. Similar reult were obtained on June 7 excet that the ignal-to-noie ratio wa by 5-1 db le than that on June 6 (ee Fig. 4.4 a and Fig b). Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15

29 STCU PARTNER PROJECT P 54 FINAL REPORT SF PAGE 6 The ectral width a a function of the heating frequency i hown in Fig.4.7 for each of 1 time interval which correond to the heating by a choen frequency. Fig Sectral width of the ignal detected in Antarctica. The ectral width at half ower wa obtained by uing moothed ectral curve. It wa found that the narrow ectral width wa about 1 H on 6/6/14 and increae to H at 6/7/14. Finally we obtain the S 4 cintillation index over 1 time interval a above. It i reented in Fig a a function of the heating frequency for June 6 and 7 exeriment. Fig.4.8. S 4 cintillation index of intenity of the detected HAARP ignal. In both exeriment the S 4 index eaked at f h =5.73 MH i.e. when the SSS, indicated by the BUM, begin to be ureed by the long cale triation uch a revealed by the DM, the latter effectively catter the HF wave. Ditribution A: Aroved for ublic relea; ditribution i unlimited. 4/5/15