You are on page 1of 11

ROUGH EVAPORATION DUCT (RED) EXPERIMENT

Kenneth D. Anderson and Richard A. Paulus* Space and Naval Warfare Systems Center San Diego, D858 49170 Propagation Path San Diego, CA 92152 Voice: 619 553 1420, Fax: 619 553 1417 kenn@spawar.navy.mil

Abstract
Prediction of electromagnetic (EM) signal propagation over a wind-roughened sea relies on a thorough knowledge of its interaction with the sea surface, the mean profiles of pressure, P, humidity, Q, temperature, T, wind, U, and their turbulent fluctuations (p', q', t', u'). Yet, within the marine surface layer, these mechanisms are not sufficiently understood nor has satisfactory data been taken to validate empirical models. The RED experiment will provide first data for validation of both meteorological and EM propagation models in the marine surface layer for rough surface conditions. Over the ocean, similarity theory is often applied to construct mean profiles of Q, T, and U in the surface layer. For smooth seas, there is good agreement between theory and measurements, both meteorological and EM. However, recent evidence indicates that even small waves perturb P, Q, T, and U profiles throughout the surface layer. Additional, although indirect, evidence of surface induced distortion has been derived from analyses of microwave signal propagation close to the sea surface. Effects of wave interaction with the turbulent fluctuations of q', t', and u' have never been studied even though q', t', and u' are crucial for modeling optical propagation. The RED experiment will be conducted offshore of Oahu, HI, from mid-August to midSeptember 2001. The Hawaiian Islands in late summer are ideal for RED because climatology shows this area and timeframe to have the highest joint probability of high winds and intense evaporation ducts (about 20 percent of the time winds are greater than 10 m s-1 with duct heights exceeding 15 m). R/P FLIP is crucial to the success of RED. Moored about 10 km off of the NE coast of Oahu, FLIP will host the primary meteorological sensor suites and serve as one terminus for the electromagnetic propagation links.

Introduction
Radar detection and tracking of low-altitude, low-cross-section targets over the sea is a function of the radar system parameters (power transmitted, frequency, etc.), target characteristics (radar cross-section, speed, etc.) and the environment (sea state, refractive conditions, etc.). Radar parameters and target characteristics may change a few dB from nominal as amplifiers age and missiles are handled but changes in environmental conditions may be responsible for many (tens) of dB changes in detection and tracking capabilities. For example, Fig. 1 (from Anderson 1995) shows the propagation loss between an X-Band radar (9.415 GHz) located 24 m above the ocean surface and a calibrated cross-section target 4.5 m above the ocean surface with respect to range. The solid curve is the propagation loss expected in a normal, or standard, refractive environment (Bean and Dutton 1968), whereas the light crosses indicate the propagation loss measured in a refractive condition of a moderate evaporation duct (Katzin, et al. 1947). At a range of 10km, which is a crucial range for weapons engagement, the radar signal is approximately 3 dB below the signal level expected in a standard atmosphere. If this radar’s detection and tracking RED BACIMO 2000 Page 1

whereas. lowaltitude targets over the sea is also keenly affected by the environment. and relating these effects to the vertical profiles through empirical “universal” functions that are stability. Both effects confuse tracking algorithms. To model EM (covering both radar and EO) system performance with any precision requires a thorough knowledge of the vertical refractivity profile. Arguably. i. carbon compounds from fossil fuel burning. ABL parameterizes the thermodynamics of the near-surface layer by incorporating the vertical wind speed profile. for EO systems. which implies a thorough knowledge of both the temperature and humidity profiles. measurements of T and Q are difficult especially as one nears the ocean surface where large water droplets with dissolved salts and suspended organics easily collect on the sensors thereby contaminating the readings. et al. uniform. Both radar and EO system performance in detecting and tracking low-altitude targets over the ocean are strongly affected by atmospheric refractive conditions. ABL theory must be correct. analogous to propagation loss for radar. Although Obukhov’s ABL parameterizations have been hugely successful for flat. is proportional to the product of the Mie efficiency factor (which is a function of wavelength. where small maybe defined as including gravity waves up to 1 m or so in amplitude.capabilities were designed for standard atmospheric conditions. t’. for a perfectly flat sea. In an analogous manner. and the latter is created by turbulent fluctuations of temperature. Over the ocean. dependent. mirages present two or more targets where only one is real and scintillation can cause targets to momentarily blink out of existence. Optical extinction. Richardson number. However.. ABL theory should provide a reasonable estimate. T. buoyancy and inertial forces.e. refractivity is dependent on the vertical distribution of mean temperature. Both molecular absorption as well as aerosol species and concentrations also affect EO systems. Extending the limit to small waves. begins to break down for wind speeds in excess of some 15 m s-1 (Edson. Significant refractive effects at EO wavelengths include mirages and scintillation where the former is created by non standard variations in the near-surface mean temperature profile. over the ocean. additional aerosols are found. particle radius and the complex optical refractive index of the particle) and the particle size distribution expressed as the number of particles per unit volume with radii between r and r+dr. 1971. For radar. Figure 2 shows the propagation factor. electro-optic (EO) detection and tracking of low-flying. as most older generation radars were. and increased concentrations of water droplets from surf generation. these measurements show that a change in environmental refractive conditions would cause this radar to miss detection of an incoming target with potentially disastrous consequences. over-land surfaces (see Businger et al. there is fairly substantial evidence that ABL theory. T. defined as the ratio of signal RED BACIMO 2000 Page 2 . Over the ocean. In littoral regions. Additional indirect evidence that Obukhov’s ABL theory breaks down under certain overocean wind speed and stability limits has been uncovered from examination of radar frequency propagation experiments. drag effects on the surface. Typically these are anthropogenic. Aerodynamic Boundary Layer theory (Obukhov 1971) has been extensively used to approximate near surface T and Q profiles with considerable success in most cases. as is water vapor in the 8 to 12 µm band (longwave). and Dyer and Bradley 1982). 1999). refractivity is dependent primarily on the vertical distribution of mean water vapor. aerosol particles consist of water drops containing dissolved salts with organics in suspension along with secondary particles such as sulfate. the parameterizations do not account for the mechanical coupling and mixing generated by ocean waves. Atmospheric CO2 is a strong absorber in the 3 to 5 µm band (midwave). Q.

2. More than three weeks of intensive observations are scheduled.3 m above the ocean and the receiver was located 4. errors approach some 15 dB. The hashed (purple) line is what is expected in a standard. 3. where theory is lacking. will have a similar vertical mast at its extreme but this mast will extend downwards into and below the ocean surface. The principal tasks of this experiment are: 1. There is good agreement between observations and predictions for the lower frequencies but.6 m above the ocean. about 17 m in length. The second task will be accomplished by analyzing this rich data set to determine regimes where current ABL theory is and is not applicable and. multi-wavelength observations of EM propagation. R/P FLIP is a stable oceanographic research platform 110 m (355 feet) in length consisting of a long slender tubular hull terminating in a normal ships bow section. sea surface elevation and wave direction. It is strongly suspected that the cause for these discrepancies is due to an improper understanding of the vertical refractivity profile. vertical distributions of aerosols. accounting for surface roughness. Mean and turbulent profiles of the basic meteorological quantities as well as ocean surface elevation statistics will be made from this mast. and boundary layers. Evaluate and validate new parameterizations. surface. The red diamonds indicate the observations. and wind speed. for a set of measurements involving 16 frequencies. is shown in Fig. refractive atmosphere. solar radiance. which is some 17 m (55 feet) in length. surface.level to the signal level expected in free space. The port side boom. The transmitter was located 1. The dashed line (green) represents the “best” calculations of propagation factor using a benchmark waveguide propagation prediction code assuming a smooth surface. Determine the extent to which ocean surface roughness modifies the vertical distribution of EM refractivity in the marine wave. in a configuration similar to what will be used for RED. The first task involves frequent and high quality in situ observations of the mean and turbulent vertical profiles of the basic meteorological quantities pressure. Surface wind speed was 8 m s-1 and the range separation was 35. surface. or normal. is hosting a one-month field experiment in Hawaii from mid-August through mid-September 2001. 1985 for a description of the “MLAYER” full waveguide propagation assessment computer code). The Space and Naval Warfare Systems Center San Diego (SSC San Diego). R/P FLIP. and boundary layers. moisture.7 nmi). of meteorological quantities and aerosol distributions in the marine wave. The RED experiment is designed to assess the effects of ocean surface roughness on both meteorological quantities and EM propagation characteristics in the marine wave. under Office of Naval Research (ONR) sponsorship. new parameterizations will be developed. The solid line (blue) represents the “best” calculations of propagation factor using the same propagation prediction code but assuming a wind ruffled surface (see Baumgartner 1983 and Hitney et al. geometries. methodologies and instruments are reviewed in the following section. Overview of the RED experiment The heart of the RED experiment is the Research Platform Floating Instrument Platform (R/P FLIP) that is operated for ONR by the Marine Physical Laboratory of the Scripps Institution of Oceanography (MPL/SIO). temperature. for the higher frequencies. and boundary layers. spanning from 2 to 18 GHz.4 km (18. Proposed platforms. Aerosols and solar radiance will be RED BACIMO 2000 Page 3 .

Q will be sensed with a chilled mirror dew-pointer (Edge Tech 2001). will sense solar heating. A receiver consisting of a 1200 mm focal length F/6 telescope. An analysis of a world wide evaporation climatology (see Patterson 1987 and Anderson 1987 for a description) included yearly joint statistics of evaporation duct height and wind speed. Southern California Greece Puerto Rico Hawaii 3.2% 11. For EO propagation measurements a wide field-of-view infrared (3 to 12 µm) blackbody source will be mounted approximately 8 m above mean sea level (msl). Fig. These units will be complimented with a sonic anemometer and an IR hygrometer (a LiCor 7500). A spectral radiometer. will provide relative 3D position information. R/P FLIP will be threepoint moored in the location shown where the water has a depth of about 370 m (1200 feet). which will also provide space for mounting EM transmitters. of up to +/. will be equipped to measure mean profiles and provide calibrations for the turbulent measurements.3% 10. The selection of Hawaii for the RED experiment came about from the need to have high evaporation ducts with high winds. mounted on the vertical mast about 5 m below mean sea level. installed at the top of the mast. Lyman-Alpha hygrometers (Mierij Meteo KOH).ten degrees. RED BACIMO 2000 Page 4 .1% 17. Sonic anemometers (Campbell Scientific CSAT3). 5 is a schematic representation of the instrumentation planned for R/P FLIP. For EO propagation. Table 1. and u'.measured by instruments close to and on the bow section. mounted near the PCASP and DMA. again. From these data. and pressure sensors (Paroscientific Digiquartz) will be mounted at multiple levels on the vertical mast to measure p'. Fig. A Trimble TansVector GPS. capable of roving from the top of the mast to the surface. located at other points on the mast. The three-point mooring will provide some rotational stability but it is expected to have azimuthal rotations. FSSP sizes radii between 1 and 30 µm and the OAP sizes between 30 and ~100 µm. Hawaii has the highest probability of evaporation duct heights greater than 15 m with wind speeds greater than 10 m s-1. A traveler. both the transmitter (source) and receiver should be low to the water with a path length of approximately 10 km. Accelerometers. The percent of time where evaporation duct heights greater than -1 15 m occurs with a wind speed greater than 10 m s (annual). Larger particles (radii from 1 to ~100 µm) will be profiled using a Forward Scattering Spectrometer Probe (FSSP) and an Optical Array Probe (OAP). Aerosols will be measured at several locations onboard R/P FLIP. provides 3D real-time position information. Smaller particles will be measured at a higher altitude using a Passive Cavity Aerosol Spectrometer Probe (PCASP) and a Differential Mobility Analyzer (DMA). These sensors will be aspirated and baffled to minimize particle contamination. q'. dual liquid nitrogen cooled detectors (HgCdTe and InSb).7% For radar frequency propagation. and 3D t'. 4 shows the planned location of R/P FLIP and both of the EM paths. the ideal path geometry is a low-sited receiver with low-sited transmitters on a path of approximately 30 km. provides sea elevation statistics and wave direction. The joint probability of such occurrences for selected locations is listed in Table 1. A Nortek 3D acoustic velocimeter. caused by wind loading on the structure. Additional surface statistics will be obtained from a standard wave wire or from an array of wave wires.

Sampling rates at the receiver will vary between a few Hertz and several hundred Hertz to obtain measurements of mean and turbulent optical transmission. To minimize logistic requirements. Doss-Hammel Prof. and 17-GHz (S-. These radar frequencies were chosen to optimize analysis of evaporation duct effects on propagation. the signal levels decrease because the dielectric surface waveguide (evaporation duct) supports multi-mode propagation. 10-. Anderson Dr. C. which can constructively and destructively combine at the receiver. RED PIs and responsibilities SSC-San Diego SSC-San Diego SSC-San Diego UC Irvine UC Irvine NPS Mr. Each transmitter will radiate 4 W (36 dBm) using separate. R. Surface-based ducts are generally much thicker than evaporation ducts.5 dB from 2 through 18 GHz) directs the signals to a HewlettPackard 8566B spectrum analyzer. J. A seventh frequency is planned to monitor possible surface-based ducting conditions that may arise from air mass advection or subsidence. The expected signal level for S-Band signals increases slowly and monotonically with increasing evaporation duct height. will be located at two altitudes. Any evaporation duct will not affect VHF signal levels at the receiver site. Ku-Band signal propagation is similar to X-Band but the signal levels increase faster and multi-mode propagation occurs for lower evaporation duct heights. approximately 4 and 12 m msl. Hristov Dr. lownoise amplifier (noise figure < 2. negating the normal requirement for a supply of helium to fill balloons. Friehe Dr. With this configuration. The radar frequency receiver consists of a 1. Principal investigators are listed in Table 2 and a schedule of R/P FLIP operations is listed in Table 3. A broadband. In addition. X-Band signal levels initially increase faster but. approximately 10 km from R/P FLIP. Radiative Transfer EO propagation Surface meteorology Wave boundary layer EM modeling RED BACIMO 2000 Page 5 . these radiosondes will aid in characterizing the surface layer up through the mixed layer. Table 2. These transmitters are crystal controlled oscillators providing short-term frequency stability on the order of 1 x 10-8. K. the radiosondes will be flown on a kite. Radar frequency transmitters. horizontally polarized. Chemistry. at some duct height (geometry dependent). each of the six transmitted frequencies can be sampled at a 0. and Ku-Band). Reid Dr. Flying radiosondes on a kite also provides for atmospheric profiling. standard gain horns with gains of about 16 dB (20 degree beamwidth). S.5 Hz rate.2 m (4 feet) diameter parabolic antenna with a broadband logarithmic feed (Tecom). Standard upper-air meteorological observations (radiosondes) will be launched from R/P FLIP to further aid in surface-based duct detection. so a 262 MHz transmitter is added to the radar frequencies. X-. Janaswamy Radar frequency propagation Aerosols. T.and data acquisition circuits will be located onshore Oahu. These tropospheric waveguides are capable of ducting VHF signals. being on the order of 100 m thick. nominally at 3-.

K. SSC San Diego Tech. 46 pp. 3 point moor..B.Table 3. Ant. References Anderson.J. M. Turbulence in an atmosphere with non-uniform temperature.R.. Proc. New York: Dover. Radio Meteorology. and E.W. and E.D. 22.A. Katzin.. 1999.D. 610 (ADA133667). R/P FLIP Schedule for Summer/Fall 2001 June July Aug Aug-Sep Sep Oct Dry Dock (San Diego) Sea trials and tow to Hawaii (Pearl) (20-21) Tow from Pearl to site. is hosting a one-month field experiment in Hawaii from mid-August through mid-September 2001. Wyngaard. 1968.B. ONR Memorandum. and boundary layers. XWVG: a waveguide program for trilinear tropospheric ducts..H. IEEE Trans. 1947. Coupled marine boundary layers and air-sea interaction initiative: Combining process studies. under Office of Naval Research sponsorship. Pappert. Patterson. Anderson.M. Bean.D. 73(2). 1987. 788-794 Dyer. 1995. A. Obukhov. Izumi.R. Worldwide distributions of shipboard surface meteorological observations for EM propagation analysis. Businger.J. Edson.. Tropospheric radio propagation assessment. Proc. J. 43(6).. Atmos. Binnian. and W.V. J. Flux-profile relationships in the atmospheric surface layer.F. Bauchman. Hitney..F. Richter. Jr. and numerical methods.. Radar detection of low-altitude targets in a maritime environment. B. Doc.and 9-centimeter propagation in low ocean ducts. H. J.L. SSC San Diego Tech. RED BACIMO 2000 Page 6 . Doc.. Bradley. 1149(ADA189157). 1983. 1982. and G. A.. Historical electromagnetic propagation condition database description.. 265-283. Bradley..A. tow to Pearl Prepare for NSF Hawaii Ocean Mixing Experiment Summary The Rough Evaporation Duct experiment is an intensive field study to assess the effects of ocean surface roughness on both meteorological quantities and EM propagation characteristics in the marine wave. and Prop. 3-19.. Baumgartner. 35(9). Doc. R. 2.L. 891-905. 30. Anderson.. The Space and Naval Warfare Systems Center San Diego. Dutton. Y. 74 pp. IEEE. et al.. 729.M. surface. simulations. 1971. 609-613. B. and E. An alternative analysis of flux-gradient relationships at the 1976 ITCE. K.C. Baumgartner. Sci. J. B.L. 206 pp. SSC San Diego Tech. K. 1987. J. R. W. 1971.E. start RED (21-13) RED (13-14) End RED.M. I. G. 1150 (ADA188771). (in preparation)... 3.

An evaporation duct is responsible for the decrease in signal (increased propagation loss) for ranges from 7 to 15 km. RED BACIMO 2000 Page 7 . Measured X-Band propagation loss (gray crosses) compared to predictions for a standard atmosphere.Figure 1.

The rough and smooth curves are modeled propagation factors assuming a rough surface (8 m s-1 winds) and a smooth sea surface respectively. dB 0 -10 -20 -30 -40 -50 -60 -70 -80 0 2 4 6 8 10 12 14 16 18 20 Frequency.20 10 Propagation Factor. Rough Smooth Standard Observed RED BACIMO 2000 Page 8 . GHz Figure 2. Multi-frequency measurements of propagation factor taken during conditions of a moderate evaporation duct with 8 m s-1 winds.

R/P FLIP at sea in a configuration similar to what will be used for the RED experiment.Figure 3. RED BACIMO 2000 Page 9 .

The approximate locations of R/P FLIP and the EM propagation paths.Figure 4. RED BACIMO 2000 Page 10 .

Figure 5. RED BACIMO 2000 Page 11 . A schematic representation of the instrumentation planned for RED.