The Effect of Radar Azimuth Angle on Cultural Data

Home , Azimuth

DR. M. LEOKARDBRYAN Jet Propulsion Laboratory Califori~iaInstitute of Technology Pasadena, CA 91 103 The Effect of Radar Azimuth Angle on Cultural Data

Differences in grey tone within adjacent areas of the Los Angeles urbanized region resulted from angle differences between the radar azimuth and the street pattern and, more importantly, the orientation of the walls of the structures imaged.

INTRODVCTIOS near (0.7 - 1.5 ,UIII) and therlnal (8- 14 pm) A HE STUDY OF URBAS AREAS fro111 gro~mcl, recent review of the use of Landsat for urban T aircrafi, and satellite obseivations is of studies is given in Carter et al. (1977). continuing interest. Land-use studies, based Generally, for the microwave poition of upon data obtained fro111 high altitudes, are the EM spectruml, a less detailed effort at data proving quite useful for observing the interpretation has heen conducted. This is dy~laniicsof urban morphology (change de- primarily because the passive (radiometer)

ABSTRACT:The use of Synthetic Aperture Radars (~AR)for tlze stzldy of cziltural and urban scenes has been prirnarily in the area of interpretation of tlze major aspects of land use and urban nlorphology. In such studies the image grey tone, distribution of bright (point) returns, and linear features have been the focus of the major efforts. It is em))hasized in this paper that tlze orientation of tlze features studied is a major factor in the grey tone of the I-est~ltingpositiue image. Thus, knowledge of this orientation, wit11 respect to the azimt~tlzangle ji.e., look-direction) of the radar antenna, is of major importance. Within the Los Aizgeles urbanized region, large areas have significantly lower grey tones than adjacent areas having similar land couer. This effect i.s determined to be the result of the angle difference between the rudur azimuth und the street pattern and, more importantly, the orientation of the walls of the structures imaged. Thus, an (1 priori knowledge ofthis orientation is necessary in order to ensure acc~irateinterpretation of the radar imagery. For radar systems operated from platforms which haue fixed azimuth angles (e.g., satellite systems ,such as Seasat-A), an interr~retation methodology, which considers the street patterns, is considered especially critical for proper and accurate interpretation of SAR imagery. tection) as well as for preparation of land-use systems generally have a very low resolution nlaps of more static scenes. Much of the re- and imagery from the active systems (radars) mote sensing work co~lcer~li~lgsuch scenes has been difficult for illany investigators to has dealt primarily with the shorter obtain. However, this last constraint has wavelengths of the electromagnetic (EM) eased considerably in the past several years, spectrum, i.e., photography (0.4 - 1.5 pn), and some effb~tshave been conducted with Landsat (0.5 - 1.1 pli~),and infrared, both respect to the iilterpretation of urban scenes

PHOTOGRAMMETRICENGINEERING AND REMOTE SENSING, Vol. 45, No. 8, August 1979, pp. 1097-1107. PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1979 from airborne radars. ,Although Seasat-A (1973) with respect to large topographic ele- (launched on 26 June 1978) failed on 9 Oc- vations in Panama and, for small features tober 1978, it had completed about 500 syn- (river bluff), by Bryan and Hall (1976). In thetic aperture radar passes of two to ten urban areas, however, where the elevations minutes duration each. With the availability of the individual features (e.g., buildings) of these data to the scientific community, it are generally much less, and especially with is expected that more investigators will show respect to the resolution ofthe radar, such an an increased interest in such work. effort would probably be of minimal utility. With respect to the operation, design, and Stereo radar also increases considerably the utility of radars for observing, monitoring, ability to determine height and slope angle, and measuring the Earth's surface, several as demonstrated by Leberl (1973), but such excellent articles are available. Jensen et al. data are even less frequent than those from (1977) present an easily readable paper single or multiple look-direction radars. Fi- oriented primarily toward discussing the nally, Coiner and Morain (1971) note for ag- conceptual operation of the radar and in- ricultural scenes that there is a ". . . ne- cluding several examples of radar imagery. cessity to account for changes in radar Brown and Porcello (1969), on the other backscatter (grey level shifts) generated by hand, provide the reader with a more tutorial angular dependencies in the scene." They paper concentrating on an introductory en- refer primarily to the depression angle ofthe gineering approach to solving the problem of radar and not to the azimuth angle, but do obtaining radar imagery with synthetic alert the interpreter to the fact that the ge- aperture systems. Both of these papers pro- ometry of the scene is of great importance vide an excellent background to scientists when dealing with radar imagery. seeking an introduction to these remote The azimuth angle is defined as the direc- sensing systems. tion, on a plane tangent to the Earth's sur- Several additional papers have shown face, in which the radar beam is pointed. some of the basic concepts of the interpreta- Thus, although we may refer to the azimuth tion of radar images of urban and other angle (also sometimes termed the look- scenes, including Bryan (1974,1975), Coiner angle, look-direction, or by Koopmans (1975) and Morain (1971), Henderson (1973), Lewis as the scan direction) with respect to the et al. (1969), and Rosenfeld and Kimberling points of the compass, this alone is of little (1977). Although these papers have dealt utility. The more important consideration is with various wavelengths and polarizations, to determine the azimuth angle with respect it is noted that, generally, the types of fea- to the orientation of the object or target tures detectable and the methodology for being illuminated by the radar energy. Again, interpretation remain fairly constant from radar interpreters have noted the increased one radar system to another. There are, how- ease with which linear features may be dis- ever, several variations of the data which can tinguished by orienting the antenna of the occur within one given (radar) system and, as radar at slightly different azimuth angles. mentioned, with the advent of the Seasat-A Generally, this requires an additional aircraft data, some of these variables may be of major pass for the antenna is, for all practical pur- importance. poses, securely mounted to the aircraft. Sev- Two of these aspects of the radar system, eral studies have been conducted to deter- which are essentially positional or geomet- mine the effect of the azimuth angle for rical in their effect on the resulting radar geologic studies (e.g., Jefferies, 1969; Dell- imagery, are the incidence angle and the wig et al., 1968, MacDonald et al., 1969). azimuth anzle.- It is often emphasized that for best results With respect to large geographic and geo- there is a need for multiple and orthogonal logic features, especially for radars operat- azimuth angles when searching the data for ing near grazing angles, the areas in shadow geologic faults (e.g., Harwood, 1967). For on the far side of a feature will be areas of no urban areas we shall demonstrate that or- (surface) information. This does not, how- thogonal look-directions (or azimuth angles) ever imply that the shadows themselves do do not necessarily produce the optimum not provide some information for, if the in- radar data. vestigator knows some of the operating pa- Eppes and Rouse (1974) have quantita- rameters of the radar (e.g., altitude, depres- tively studied the problem of the effect of sion angle), the height of the feature may be azimuth angle variations on the detectability determined from the length of the shadows of surface features. They dealt exclusively (LaPrade and Leonardo, 1969). This has with linear features which have some topo- been demonstrated by Lewis and Waite graphic expression. The results of their labo- THE EFFECT OF RADAR AZIMUTH ANGLE ON CULTURAL DATA ratory study indicate that proper spatial fil- of the ten north-south data runs shown on tering in the Fourier plane of the radar Figure 1. Density data derived from the image having a single azimuth angle could mosaic, which is composed of densitomet- provide a detectability of the linears equiv- rically and geometrically corrected SAR data alent to that obtained by several sets of radar (Bryan et al., 1977), were used for this data taken at multiple azimuth angles. Obvi- present study. Within this mosaic a series of ously, this approach can effect a consider- target areas of residential and commercial able savings in data collection and, although land use was used to test the effect of radar increasing processing costs, the overall re- azimuth angles on the resulting image tone. sult will be a lower expenditure for essen- The following discussion describes the re- tially the same information. sults of this study. To date there has been little interest in the Large sections of the Los Angeles urban radar interpretation and application commu- area (cover) have decidedly different grey nity to continue the approach initiated by tones. By concentrating on those areas which Eppes and Rouse (1974). However, the need are of the mid-grey tones, thus discounting for a revitalization may be imperative due to the very dark (gravel pits, airports, flood the type of data which are to become avail- control areas) and the very bright (industrial able in the near future from Seasat-A and areas), we note that the tonal changes are Shuttle Imaging Radar-A (SIR-A). generally well-defined, have geometric patterns with linear boundaries, and are many resolution cells in size. The cause of the formation of the bound- A set of SAR data of the Los Angeles area aries and the included grey areas is, then, an was obtained on 25 May 1977 as part of a interesting aspect of the imagery to be con- joint EuropeaniUnited States aircraft Space sidered. Changes in the land cover across Shuttle simulation program. The radar sys- the boundaries seen on the SAR imagery are tem used was the NASA L-band (1.215 Ghz) not apparent. All of the study sites which radar operated by the Jet Propulsion Labo- will be discussed were composed of resi- ratory and mounted on the NASA CV-990 air- dential andior small co~nmercial(2-3 story) craft. Only HH polarization and 20 m reso- structures. The only contrast detected lution radar imagery are discussed. Figure 1 through auxiliary information (e.g., ground shows the flight tracks of the aircraft when truth, maps, aerial photography), which cor- obtaining the radar data over Los Ange- relates with the linear boundaries ofthe grey les. The cover figure is a mosaic composed areas as seen on the SAR data, is that of the street patterns. Thus, although the street patterns within the Los Angeles urbanized area are generally rectilinear, not all are oriented in the same direction with respect to true north. This is a result of several his- torical factors, including (1) the development of individual towns around isolated centers and their growth into a complex of suburbs, (2) the street pattern following the original rancho boundaries, and, in several cases, (3) the fitting of streets to the topogra- phy. Consequently, by using radar data obtained at a constant azimuth angle and not- ing the changes in the grey tones, we can identify those areas in which the azimuth angle of the radar is critical for the change in grey tone. Large-scale vegetation has been previously noted to be an important factor in the radar backscatter of urban areas (Bryan, 1974, 1975). That effect is primarily to mask the various point targets. Consequently, the result is the smoothing of some ground de- tail. Although this also occurs in the Los FIG. 1. Flight tracks flown by NASA CV-990 Angeles data under discussion (because while collecting Los Angeles data. 25 May 1977. large areas and patterns are being studied), PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1979 the effect of vegetation is not p-ronounced. Also, because the radar system used has a 25 m resolution, the detailed effect of vegetation has been reduced.

Six sites, labelled A to F and listed in Table 1, were selected for this study. Site A: (Figure 2, Burbank). In this scene the majority of the streets are oriented in three directions, at 044O, 06B0, 000" (and orthogonally) although there are some variations such as Clybourn Avenue (340" and 350"). The dominant land use is single family residential with major strip commercial areas along Burbank Boulevard, Magnolia Boulevard, and Olive Avenue. The strongest returns are from areas where the streets are oriented true north-south or east-west. The FIG.2. Study Site A - Burbank, California. Radar radar orientation is 090" (Table 1).The ex- azimuth is 090" true. tremely bright point return in the south cen- (a) L(HH) SAR image; (b) Sketch map of street tral portion ("D" on Figure 2) is from the patterns. buildings of the Disney Studio which are B: Burbank Stud~os oriented north-south within an area where D: Disney Studios all other buildings and streets are skewed to LGCH: Lakeqide Golf Course of Hollywood the north-south grid. Freeways and major NHP: North Hollywood Park streets appear as dark (no return) areas. SJMC: Saint Joseph Medical Center Site B. This area is quite similar to Site A with respect to street pattern and radar return. In the areas of darker tones the two major street orientations are 021" and 042" (and their orthogonal directions). The light toned areas have streets oriented to the cardinal points. Site C: (Figure 3, Sun Marino, Sun Ga- briel, Temple City). For the first two examples, the radar azimuth was 090"; for Site C this angle has been rotated to 26g0. Again, those areas with streets oriented to the cardinal voints of the comDass are high- return areas,and areas within which the streets are

Radar Aircraft Azimuth Site Run Track Angle Location A 12 000" 090" Burbank B 8 000" 090" Altadena 7 179" 269" San Marino & San Gabriel FIG.3. Study Site C - Alhambra, San Gabriel, San Marino, and Temple City, California. Radar D 11 18%" 278' Los Angeles azimuth is 269' true, E 11 188O 278' Inglewood (a) L(HH) SAR image; (b) Sketch map of street patterns. F 7 179" 269' Pico Rivera & Whittier AMCC: Alhambra Municipal Golf Course SGCC: San Gabriel Country Club THE EFFECT OF RADAR AZIh4UTH ANGLE ON CULTURAL DATA not parallel or orthogonal to the radar azimuth angle have significantly lower return. The dark grey area to the east is a portion of Temple City in which the streets are oriented at 075@,07QU, and orthogonally. The boundaries of this area are Duarte Avenue, Baldwin Avenue, Encinita Avenue, and Longden Avenue. The southern boundary of this grey zone is quite diffuse because there I are very few streets in this area, which is composed primarily of large storage yards and light industry (in the vicinity of the Southern Pacific Railroad and Eaton Wash). b In the west portion of the study site is a similar grey area composed primarily of portions ofthe Alhambra and San Marino politi- cal divisions. Botmdaries here are Garfield i and Atlantic Avenues (to the west) and the Southern Pacific Railroad (to the south). The eastern boundary follows a series of inter- connecting street patterns, Radar azimuth is 278' true. Two golf courses, the Alhambra Munici- patterns. pal and the San Gabriel Country Club, are easily identified as areas of very low returns EP: Exposition Park within a zone of cardinally oriented streets. Also of major interest are the bright linears which are portions of Eaton Wash. This is a normally dry wash (25 feet wide and 10 feet East of this, six areas of alterhating street deep) with vertical concrete walls and bor- patterns are clearly seen on b th the radar dered by a chain link fence. This feature is imagery and the street pattern haps. easily followed where it is paralIel to the The area between Figuero Street and flight track (and therefore normal to the radar Main Street consists primarilyt of low (2-3 azimuth direction). As in the case of the rail- story) commercial structures. $ roads, it can often be traced as a dark linear Street residential units due to the relatively broad right-of-way family buildings and through the majority of the San Gabriel por- units predominate. Also, for th two blocks tion of the image. south of the Santa Monica ~rerwayand for Site a: (Figure 4, Los Angeles). This scene, from the central portion of Los Angeles and centered on the University of Southern California (USC)and Exposition Park (EP), provides an opportunity to note the fine degree to which the street patterns can be identified using the L-band, 20 m resolution radar. USC and Exposition Park are both large dark grey areas on the image with individual buildings providing some point returns within these dark zones. East of Ex- position Park and north of Santa Barbara Street is an area of major interest. For these data the aircraft track was 188", the radar azimuth was 278", and the radar was oriented off the normal with respect to the north- south oriented street grid and approximately 12" off normal with respect to the majority of the other streets, The small area bounded by Figueroa Street and the Harbor Freeway, an area of north-south streets, is only two city thogonal to the radar azimuth. En the central blocks wide but is of sufficient width to appear as a bright pattern on the SAR imagery. PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1979

FIG.5. Study Site E - Inglewood, South Los Angeles, California. Radar azimuth is 278" true. (a) L(HH) SAR image; (b) Sketch map of street patterns. CP: Centinela Park HP: Hollywood Park IPC: Inglewood Park Cemetery FIG.6. Study Site F - Pico Rivera and Whittier, California. Radar azimuth is 269" true. meeting at Park Circle) are approximately (a) L(HH) SAR image; (b) Sketch map of street one city block wide and produce areas of low patterns. returns. The large Inglewood Park Cemetery PP: Pico Park and Hollywood Park race track are areas of WC: Whittier College no streets and, consequently, very low returns. Also, the small area centered on Spruce Avenue and that area to the north of sult of the street pattern. Here a strip com- Centinela Park both appear similar on the mercial development in the midst of an area radar image tone. However, the land use is of rectangular street patterns oriented ap- quite different, with the former being small proximately 40" from the radar azimuth angle commercial structures and parking lots and causes the higher returns. the latter composed primarily of single fam- To the west of the San Gabriel River, iso- ily residential units. Thus, this portion of the lated bright returns in Pico Rivera are again image emphasizes the dominance of the the result of small areas of streets parallel to street pattern and the decreased depen- the radar azimuth angle. The extremely dence of the tone of the L-band radar image bright return northwest of the comer of Pas- on the land cover. sons Blvd. and Washington Blvd. is the El Site F: (Figure 6, Pico Rivera, Whittier). Rancho High School complex in Pico Rivera. The right angle grid system of the City of These are oriented in a north-south direction Whittier, oriented to the cardinal points of and are, therefore, parallel to the aircraft the compass, is clearly outlined by several flight track. streets (Norwalk Blvd., Whittier Blvd., In Whittier immediately west of Whittier Washington Blvd., College Ave.) and, to the Blvd. and north of Washington Blvd. a bright north, the Puente Hills. Several small areas north-south linear is a small access road to where canyons and the park-like atmosphere the light industries. To the west are large of Whittier College ("WC" on Figure 8) dis- relatively open areas which are dark on the rupt the regular grid pattern and are defined radar imagery. on the radar image as areas of very low re- The bright returns between Washington turn surrounded by the higher return of the Blvd. and Rivera Rd. have a few north-south parallel and orthogonal streets, oriented streets with many commercial in- In the central portion of the image and dustrial building oriented parallel to the west of the intersection of Hadley St, and streets. However, between Rivera Rd, and Whittier Blvd., the high return is not the re- Slauson Ave. a few alleyways exist for access THE EFFECT OF RADAR AZIMUTH ANGLE ON CULTURAL DATA

against theta (defined as the angle between the radar azimuth and street orientation, as referenced to true north) (Figure 7) is presented. In this figure it is noted that, for areas where theta is less than 10", the radar image is bright (116 to 193; average s152). In areas where theta is greater than 10' the image brightness is considerably lower (93 to 128; average =115). The critical value for theta at which this brightness changes is between approximately 10' and 15". This critical value of theta is also independent of the land use (with respect to the two broad catagories previously defined). The imaging radar used is designed to simulate Space Shuttle and Seasat-A SAR and provide pre-launch engineering and data interpretation experience and guidance. The Seasat-A incidence angle was 20" (off nadir) THETA whereas for the airborne radar the incidence FIG. 7. Plot of grey value of SAR image versus angle is 0"- 45". We have, then, the possibil- theta, the angle between the radar azimuth and ity that the transfer of interpretation experi- the street orientation. ence from the aircraft to the spacecraft data : theta < 10" may be confused by this change in the de- 0 : theta > 10" pression angle. However, because no effect of range is seen with respect to theta on the aircraft data, it is believed that Seasat-A data to buildings which are oriented approxi- will show a similar lack of range effect. Pre- mately 20" off the north-south lines. liminary inspection of Seasat-A SAR data Finally, on this image, it is noted that in show this to be indeed the fact. the central part of Whittier (in the vicinity of There is the possibility that this azimuth Milton Ave. and Hadley St.) the effect of the angle effect described may be unique to the street pattern is seen in the strong return in L-band radar and at this resolution, although this area of small shops and commercial ser- we are not able at this time to adequately test vices. Also, to the north and west (toward the this hypothesis. It is recalled that, in earlier intersection of Beverly Blvd, and Milton papers discussing L-band radar (A - 21 cm, Ave., and beyond) the land use changes but resolution of 30 ft) (e.g., Bryan, 1975), it was the street pattern continues to dominate the noted that, for commercial, industrial and radar return. residential areas, land use could be inter- preted quite accurately even by inexperi- enced interpreters. In those data no in- stances were observed which related the The lack of correlation of the actual land effect of street orientation and radar azimuth use (low commercial and residential units) as described in the present paper. To further and the radar image tone is apparent. These test this hypothesis a set ofdata, collected by urban data and previous works have em- systems operating at the same wavelength phasized that the angles at which the radar but at different resolutions for the same energy strikes a reflecting surface is a major study site, needs to be available. Such data control of the backscatter (brightness) of the are not presently available. radar image. We are not concerned here with Imagery at X- and L-band, and HH and the other properties of the surfaces such as HV polarizations, are available, however, for surface roughness, dielectric constant, con- Sun City, Arizona, and permit comparisons struction materials, vegetation cover, etc., all of a rather unique urban morphology. In of which have a definite effect on the radar Figure 8, an X-band image (Goodyear, backscatter. What it interesting is the appar- ANIUPD-4) of Sun City, Arizona is pre- ent consistency of the relationship between sented. Interestingly, we note in the circular the radar azimuth angle, the street pattern, patterns of the streets two sets of high re- and the change in overall grey tone. turns. These high returns are located at or In order to test this effect and to define near tangents to the street pattern which are the angle, a graph plotting the grey level nearly normal or parallel to the radar PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1979

lationship of street direction versus azimuth angle.

AN EXPLANATION 4 Are street patterns being detected and, if so, why, on the original imagery of Los Angeles, are streets per se often not visible? This would seem to indicate that something other than street patterns is imaged, and that the effect under discussion is evidently caused by features highly correlated with the street pattern. These features may be related to the orientation of the buildings and their walls, which are normally related FIG.8. X-band (HH) image obtained by Good- year Aerospace using an ANIUPD-4 radar. Sun to the street patterns. In many areas of west- City, Arizona. Note the locations of the high re- ern society, and in relatively modern towns, turns within the circular portions of this urban a common practice is to construct rectangu- scene. lar buildings parallel to the streets. If the street is curved, the practice is to place the buildings with one side parallel to the tan- azimuth direction. For the same area of Sun gent of the street curve. Hence, the situa- City, Arizona, both X- and L-band data, each tions depicted in Figures 10 and 11 arise. in two polarizations, are available. This lat- From the earlier discussion we would ex- ter set of data (Figure 9), although some- pect that a radar image obtained with the what skewed during the optical correlations, antenna located at position 1 or 2 (Figure 10) is especially interesting. Here both the X- would give a very bright return, and low re- and L-band HH data show this bright return turns would be obtained with an antenna at with respect ot the azimuth direction of the position 3 or 4. This would be true if either radar as related to the tangent to the street the orientation of streets or the building orientations. However, the HV data do not walls is the basis fbr the arguments. Ifa set of show this relationship. ideally circular streets is used as a model We thus have an indication that the polari- (Figure 11) a radar image from a position of zation may also be a crucial factor in this re- either 1 or 2 should have bright returns. But,

L-HH

X-HV 1-HV _ I. ' FIG.9. Synthetic aperture radar imagery at two wavelengths and two polarizations. Sun City, Arizona. Data collected by the Environmental Research Institute of Michigan, 05 April 1974. THE EFFECT OF RADAR AZIMUTH ANGLE ON CULTURAL DATA

For the Los Angeles residential areas, considering that the radar resolutioll is 20 nl, if the street patterns caused the bright returns, it would be expected that the high returns would be concentrated along the actual streets. However, the high returns are rather consistantly distributed across large areas. Also, the streets which are smooth and normally imaged as dark (specular returns) would not themselves cause the high returns. Thus, the building walls, forming di- hedral reflectors, cause the high backscatter. The depressioll angle of the antenna (or the incidence angle to the ground surface) has little impact on this palticular type of return, and the azimuth angle is the controlling fac- MB tor. The exact explanation of the critical FIG. 10. Positions of buildings in a rectilinear azimuth angle (theta) of 10" - 15" has not yet street grid. been identified; it is presently believed, however, that this is an effect of the radar as we note in the SLIICity data, such is not antenna beamwidth. I the case. In Figures 8 and 9 the antenna azimuths on the strong returns are oriented toward the center of the circular pattern and independent of the straight streets. Those It has often been argued, and rightfully so, portions of the streets (curbs) which might that a set of multiple passes using an imaging be available for the bright reflections are of radar will enhance the accuracy of the in- insufficient physical size to yield the bright terpretation of the radar data obtained. returns. Therefore, the only remaining items MacDonald et al. (1969) suggest that as aligned with the observed bright returns many as eight should be made to provide the would be the walls of the buildings, which optimum data \et. However, this inany pass- are arranged with respect to the streets as es may be redundant. If some prior knowl- indicated in Figure 11. This conclusion is edge about the orientation of the features to further supported by the data shown in Fig- be studied is available, then a set of flight ure 2. In this figure the Disney Studios, con- tracks which will be oriented to take advan- sisting of several large buildings oriented in tage of this a priori knowledge can be desig- a north-south direction (normal to the an- nated. Assuming that a major interest is to tenna orientation), cause very bright returns. define these linear features rather than to Hence, it is the walls of buildings which identify the nature of each side of the feature cause the bright returns. (e.g., two sides of a mountain range) and that radar shadow is not pronounced, imagery from opposite azi~lluthdirections is redundant. Imagery obtained with the antenna azimuth normal to the linear trend and at some angle of between 15" - 75" from that normal should be sufficient. Thus, knowledge of the flight track is quite important for radar image interpreters. Although there are the possibilities of obtaining Seasat-A data from two different azimuth angles (generally in the north-east and north-west look- directions with an ascending and descend- ing orbit for mid-latitude locations), these angles may not be optimal for change detection of the land-use patterns. In the instant " 2 case of Los Angeles, the same areas inay be

ME continually imaged as bright or dark, and the +- effect of azimuth angle may mask some FIG.11. Positions of buildings in a circular street changes in land use. Thus, computer clas- grid. sification of land use may be difficult. This 1106 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING. 1979

is, indeed, a problem which is siinilar to that pulsion Laboratory, California Institute of encountered by users of Landsat data in Technology, under contract NAS7-100, which, although the timing is consistent, sponsored by the National Aeronautics and there is a seasonal change of sun angles for Space Administration. each succeeding observation of a given The computer processing and mosaicking scene. Possibly, fbr the study of radar imag- of the Los Angeles imagery was conducted ery, an approach as initiated by Eppes and by W. Stromberg; S. Scott did the darkroom Rouse (1974) would be a logical, initial point processing and L. Shelton executed most of to decrease the dependence of the radar re- the cartography. E Petersen and the crew of turn on azimuth angle.- When accurate in- the CV-990 are also thanked for their de- formation concerning the linearity and tailed efforts at determining the flight lines orientation of the features on the Earth's and executing the data collection. D. Billeu, surface is available, it should be a relatively working under some very stringent condi- simple problem to use filters to subdue or tions, operated the SAR during the collection remove this effect. If we are to attempt this of the data. approach in the interpretation of satellite radar data, we may encounter limited application for some older or non-western towns. Brown, W. M., and L. 1. Porcello, 1969. An Intro- As pointed out by Carter et al. (1977), duction to Synthetic Aperture Radar, IEEE "There is also (in the United States) a very Spectrum, V. 6, Sept., pp. 52-62. marked grid systein of transportation net- Bryan, M. L., 1974. Extraction of Urban Land works. Such geometrical simplicity is not Cover Data from Multiplexed Synthetic usual in the United Kingdom." Nor is it true Aperture Radar, Proc. 9th International Sym- posium on Remote Sensing of Environment, in all of the U.S. (e.g., New England). In- Ann Arbor, MI. pp. 271-288. deed, both of the test cases presented here 1975. Interpretation of an Urban Scene represent quite young towns. Using Multi-Channel Radar Imagery, Re- mote Sensing of Environ., V. 4, No. 1, pp. 49-66. Bryan, M. L., and D. K. Hall, 1976. A Comparative This study of 25 cm (L, HH), 20 m resolu- Study of Active and Passive Microwave Im- tion radar data has indicated that the differ- agery Over the North Slope of Alaska, Proc. ence between the azimuth angle of the radar Association of Amer. Geographers, V. 8, pp. and the orientation of the streets (or other 164-168. linear features) is of critical importance. For Bryan, M. L., W. D. Stromberg, and T. G. Fan, the Los Angeles area, the angle at which the 1977. Computer Processing of SAR L-Band change in the backscatter is most pro- Imagery, Photo. Eng. and Remote Sensing, V. 43, No. 10, Oct., pp. 1283-1294. nounced is in the neighborhood of 10" to 15'. For this study area radar range (depression Carter, P., W. E. Gardner, T. E. Smith, and M. J. Jackson, 1977. An Urban Management Infor- angle or incidence angle) is apparently of mation Service Using Landsat Imagery, minor importance in controlling the Photo. Record, V. 9, No. 50, Oct., pp. 157-171. backscatter from flat urban areas having Coiner, J. C. and S. A. Morain, 1971. Image In- numerous vertical components in the form of terpretation Keys to Support Analysis of walls. It is determined that an awareness of SLAR Imagery, Proc. Am. Soc. Photo., Fall the linearities and the orientation of the Conv., Sept. pp. 393-412. cultural features will be critical for an accu- Dellwig, L. F., J. N. Kirk, and L. H. Jefferies, rate interpretation of radar data from either 1968. The Importance of Radar Look- aircraft or spacecraft. Thus processing tech- Direction in Lineament Pattern Detection, niques need to be developed to remove this Abstracts, Geological Soc. of Amer. Annual effect prior to interpretation of an urban area Meet., Mexico City, Nov., 1968, pp. 72. for land-use or change detection investiga- Eppes, T. A., and J. W. Rouse, Jr., 1974. tions. With respect to the satellite data, this Viewing-Angle Effects in Radar Images, need may be more critical than for aircraft Photo. Eng., V. 40, No. 2, Feb., pp. 169-173. data because in the latter case it is possible Harwood, D. S., 1967. Radar Imagery: Par- to change the vehicle flight track to compen- machenee Lake Area, West-Central Maine, sate for the orientation of features which are Washington, D.C., USGS Earth Resources Survey Program, Tech. Let. NASA-81, June, of particular importance and interest. 6 pp. + figs. Henderson, F. M., 1973. Small Scale Land Use Mapping with Radar Imagery, Lawrence, Kansas, Ph.D. Dissertation, U. Kansas, Geog- This paper presents the results of one raphy, ix, 164 pp. (Ann Arbor, Michigan, phase of research carried out at the Jet Pro- Univ. Microfilms #74-12-571). THEEFFECTOFRADARAZIMUTHANGLEONCULTURALDATA

Jefferies, L. H., 1969. Lineaments in the Grand Simonett, 1969. Detection of Linear Cultural Canyon Area, Northern Arizona - A Radar Features with Multipolarized Radar Imagery, Analysis, Lawrence, Kansas, U. Kansas, Proc. 6th International Symp,osium on Re- CRES, Inc., Report 118-9. ii, 16 pp. (Available mote Sensing of Environment, Ann Arbor, NTIS N69-32799). MI., pp. 879-893. Lewis, A. J., and W. P. Waite, 1973. Radar Shadow Jensen, Homer, L. C. Grahm, L. J. Porcello, and Frequency, Photo. Eng., V. 38, No. 2. Feb., pp. E. N. Leith, 1977. Side-Looking Airborne Ra- 189-196. dar, Scientijk American, V. 237, No. 4, Oct. pp. 84-95. MacDonald, H. C., et al., 1969, The Influence of Radar Look-Direction on the Detection of Koopmans, B. N., 1975. Variable Flight Parame- Selected Geological Features, Proc., 6th In- ters for SLAR, Photo. Eng., V. 41, No. 3, Mar., ternational Symposium on Remote Sensing of pp. 299-305. Environment, Ann Arbor, MI., pp. 637-650. LaPrade, G. L., and E. S. Leonardo, 1969. Eleva- Rosenfield, C. L., and A. J. Kimerling, 1977. tions from Radar Imagery, Photo. Eng., V. 35, Moving Target Analysis Utilizing Side- No. 4, Apr. pp. 366-371. Looking Airborne Radar, Photo. Eng. and Re- Leberl, Franz, 1973. Radargrammetry for Image mote Sensing, V. 43, No. 12, Dec., pp. 1519- Interpreters, Enshede, ITC Lecture Notes, 1522. 136 pp. (Received September 1, 1978; revised and ac- Lewis, A. J., H. C. MacDonald, and D. S. cepted March 12, 1979)

First Announcement l5P 14th International Congress HAMBURG

of the 1980

International Society for Photogrammetry

Hamburg, Federal Republic of Germany July 13-25, 1980

The 14th International Congress, to be held in the Congress Centrum hamburg, will include technical conferences, exhibits, technical tours, excursions, and social events. For further information please write to The Secretariate ISP Congress 1980 c/o Hamburg Messe und Congress GmbH Congress-Organisation P.O. Box 30 23 60 D-2000 Hamburg 36 Federal Republic of Germany