• Able, K. P., 1972: Fall migration in coastal Louisiana and the evolution of migration patterns in the gulf region. Wilson Bull.,84, 231–243.

  • Berthold, P., 1996: Control of Bird Migration. Chapman and Hall, 355 pp.

  • Crum, T. D., and R. L. Alberty, 1993: The WSR-88D and the WSR-88D operational support facility. Bull. Amer. Meteor. Soc.,74, 1669–1687.

    • Crossref
    • Export Citation
  • ——, ——, and D. W. Burgess, 1993: Recording, archiving, and using WSR-88D data. Bull. Amer. Meteor. Soc.,74, 645–653.

    • Crossref
    • Export Citation
  • Duncan, R. A., 1994: Bird migration, weather and fallout: Including the migrant traps of Alabama and northwest Florida. 95 pp. [Available from R. A. Duncan, 614 Fairpoint Dr., Gulf Breeze, FL 32561.].

  • Eastwood, E., 1967: Radar Ornithology. Methuen, 278 pp.

  • Fisher, N. I., 1993: Statistical Analysis of Circular Data. Cambridge University Press, 277 pp.

    • Crossref
    • Export Citation
  • Forsyth, B. J., and D. James, 1971: Springtime movements of transient nocturnally migrating landbirds in the gulf coastal bend region of Texas. Condor,73, 193–207.

    • Crossref
    • Export Citation
  • Gauthreaux, S. A., Jr., 1969: A portable ceilometer technique for studying low-level nocturnal migration. Bird-Banding,40, 309–320.

    • Crossref
    • Export Citation
  • ——, 1970: Weather radar quantification of bird migration. BioScience,20, 17–20.

    • Crossref
    • Export Citation
  • ——, 1971: A radar and direct visual study of passerine spring migration in southern Louisiana. Auk,88, 343–365.

    • Crossref
    • Export Citation
  • ——, 1972: Behavioral responses of migrating birds to daylight and darkness: A radar and direct visual study. Wilson Bull.,84, 136–148.

  • ——, 1992: The use of weather radar to monitor long-term patterns of trans-Gulf migration in spring. Ecology and Conservation of Neotropical Migrant Landbirds, J. M. Hagan III and D. W. Johnston, Eds., Smithsonian Institution Press, 96–100.

  • ——, D. S. Mizrahi, and C. G. Belser, 1998: Bird migration and bias of WSR-88D wind estimates. Wea. Forecasting,13, 465–481.

    • Crossref
    • Export Citation
  • Haro, J., and S. Gauthreaux, 1997: WSR-88D VAD wind profile data influenced by bird migration over the southwest United States. NOAA Tech. Memo. NWS WR-244, 26 pp. [Available from National Technical Information Service, U.S. Dept. of Commerce, 5285 Port Royal Rd., Springfield, VA 22161.].

  • Kerlinger, P., 1989: Flight Strategies of Migrating Hawks. University of Chicago Press, 375 pp.

  • ——, and F. R. Moore, 1989: Atmospheric structure and avian migration. Current Ornithology, D. Powers, Ed., Vol. 6, Plenum Press, 109–142.

    • Crossref
    • Export Citation
  • Klazura, G. E., and S. A. Imy, 1993: A description of the initial set of analysis products available from the NEXRAD WSR-88D system. Bull. Amer. Meteor. Soc.,74, 1293–1311.

    • Crossref
    • Export Citation
  • Lack, D., and G. C. Varley, 1945: Detection of birds by radar. Nature,156, 446.

    • Crossref
    • Export Citation
  • Larkin, R., 1984: The potential of the nexrad radar system for warning of bird hazards. Proc. Wildlife Hazards to Aircraft Conf. and Workshop, Charleston, SC, FAA, 369–379. [Available from Dept. of Transportation, Federal Aviation Administration, Office of Airport Standards, Washington, DC 20591.].

  • Lowery, G. H., Jr., 1951: A quantitative study of the nocturnal migration of birds. Univ. Kans. Publ. Mus. Nat. Hist.,3, 361–472.

  • ——, and R. J. Newman, 1955: Direct studies of nocturnal bird migration. Recent Studies in Avian Biology, A. Wolfson, Ed., University of Illinois Press, 238–263.

  • Nisbet, I. C. T., 1959: Calculation of flight directions of birds observed crossing the face of the moon. Wilson Bull.,71, 237–243.

  • ——, 1961: Studying migration by moon-watching. Bird Migr.,2, 38–42.

  • ——, 1963a: Estudio de la migracion sobre el disco lunar (I). Ardeola,8, 5–17.

  • ——, 1963b: Quantitative study of migration with 23-centimetre radar. Ibis,105, 435–460.

    • Crossref
    • Export Citation
  • Richardson, W. J., 1990: Timing of bird migration in relation to weather: Updated review. Bird Migration: Physiology and Ecophysiology, E. Gwinner, Ed., Springer, 78–101.

    • Crossref
    • Export Citation
  • ——, 1991: Wind and orientation of migrating birds: A review. Orientation in Birds, P. Berthold, Ed., Birkhäuser Verlag, 226–249.

  • Russell, K. R., and S. A. Gauthreaux Jr., 1998: Use of weather radar to characterize movements of roosting purple martins. Wildl. Soc. Bull.,26 (1).

  • Russell, R. W., and J. W. Wilson, 1996: Aerial plankton detected by radar. Nature,381, 200–201.

    • Crossref
    • Export Citation
  • Schaefer, G. W., 1976: Radar observations of insect flight. Insect Flight, R. C. Rainey, Ed., Blackwell, 157–197.

  • Williams, T. C., J. M. Williams, L. C. Ireland, and J. M. Teal, 1977:Autumnal bird migration over the western North Atlantic Ocean. Amer. Birds,31, 251–267.

  • View in gallery

    WSR-88D base reflectivity images (0.5° elevation) depicting a transgulf migration of birds arriving at Lake Charles, LA, 30 April/1 May 1995.

  • View in gallery

    WSR-88D base reflectivity images (0.5° elevation) depicting the departure of nocturnal bird migrants from stopover sites around Lake Charles, LA on 1 May 1995. Note the large precipitation echo moving into the surveillance area to the north of the station (c)–(f).

  • View in gallery

    WSR-88D base velocity images (0.5° elevation) depicting a transgulf migration of birds arriving at Lake Charles (a) on 30 April 1995 and (b) the departure of nocturnal bird migrants from stopover sites around Lake Charles on 1 May 1995.

  • View in gallery

    (a) Surface conditions and (b) 850-mb winds aloft on 30 April 1995.

  • View in gallery

    (a) Surface conditions and (b) 850-mb winds aloft on 1 May 1995.

  • View in gallery

    WSR-88D VAD depicting wind profiles at Lake Charles 0001-0059 UTC, 1 May 1995.

  • View in gallery

    WSR-88D base reflectivity images (0.5° elevation) depicting annular signatures associated with morning departures of purple martins from roost sites. Images show ring echoes in both clear air mode (upper panel, 8 July 1995) and precipitation mode (lower panel, 9 July 1995).

  • View in gallery

    Relationship between number of birds crossing a 1.6 km line h−1 (migration traffic rate) and the maximum reflectivity (dBZ) detected by the WSR-88D.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 274 274 25
PDF Downloads 122 122 17

Displays of Bird Movements on the WSR-88D: Patterns and Quantification

View More View Less
  • 1 Department of Biological Sciences, Clemson University, Clemson, South Carolina
© Get Permissions
Full access

Abstract

The WSR-88D can readily detect birds in the atmosphere in both clear air and precipitation mode, and echo reflectivities of 30–35 dBZ may be realized during heavy migration events or when birds are departing a roosting site. This paper describes the appearance of birds on base reflectivity, base velocity, and velocity azimuth display wind profile products, and presents a calibration curve that relates decibel values of reflectivity to bird migration traffic rates. The recognition of bird displays in WSR-88D products is essential for the accurate interpretation of data gathered by the radar and its use in the development of forecasts. The findings also document the importance of the WSR-88D as a remote sensing tool for biological studies of birds and insects in the atmosphere and the application of such information in the avoidance of bird–aircraft collisions.

Corresponding author address: Dr. Sidney A. Gauthreaux Jr., Department of Biological Sciences, Clemson University, 132 Long Hall, Box 341903, Clemson, SC 29634-1903.

Email: sagth@clemson.edu

Abstract

The WSR-88D can readily detect birds in the atmosphere in both clear air and precipitation mode, and echo reflectivities of 30–35 dBZ may be realized during heavy migration events or when birds are departing a roosting site. This paper describes the appearance of birds on base reflectivity, base velocity, and velocity azimuth display wind profile products, and presents a calibration curve that relates decibel values of reflectivity to bird migration traffic rates. The recognition of bird displays in WSR-88D products is essential for the accurate interpretation of data gathered by the radar and its use in the development of forecasts. The findings also document the importance of the WSR-88D as a remote sensing tool for biological studies of birds and insects in the atmosphere and the application of such information in the avoidance of bird–aircraft collisions.

Corresponding author address: Dr. Sidney A. Gauthreaux Jr., Department of Biological Sciences, Clemson University, 132 Long Hall, Box 341903, Clemson, SC 29634-1903.

Email: sagth@clemson.edu

1. Introduction

Since the discovery in the early 1940s that birds were responsible for some of the “spurious” echoes displayed on British surveillance radar (Lack and Varley 1945), radars have proven to be useful tools for the detection, monitoring, and quantification of bird movements in the atmosphere (Eastwood 1967). In 1957 the United States established a national network of weather surveillance radars that used the WSR-57 radar: a 10-cm (S-band) system with a transmitter power of 500 kW. This system and subsequent models (e.g., WSR-74C) proved to be extremely valuable for quantitative studies of bird migration (Gauthreaux 1970), and for three and a half decades the WSR-57 and WSR-74C were used by several ornithologists to study bird migration (e.g., Gauthreaux 1971; Able 1972; Gauthreaux 1972; Williams et al. 1977; Gauthreaux 1992).

The WSR-57 and WSR-74C radars in the national network have now been replaced with a new Doppler radar, the WSR-88D, or next generation radar (NEXRAD). These new units are markedly different from the old WSR-57 and WSR-74C units (Crum et al. 1993; Klazura and Imy 1993; Crum and Alberty 1993) and have an enhanced capability of detecting weak reflectors in the atmosphere such as birds and insects (Larkin 1984). In this paper we demonstrate that the WSR-88D can provide unmatched opportunities for gathering detailed information on many aspects of the flight behavior of birds during long-distance migrations and local movements. We discuss some of the features of the WSR-88D as they relate to radar ornithology and the appearance of birds in base reflectivity, base velocity, and velocity azimuth display (VAD) wind profile (VWP) products, and demonstrate how the radar can be used to delimit important migration stopover areas and large roosting concentrations of birds. We present a calibration curve that relates different decibel values of reflectivity of bird displays on base reflectivity products to the actual numbers of birds in the atmosphere (bird migration traffic rates). Because migrating birds may“contaminate” wind estimates by the WSR-88D (Gauthreaux et al. 1998), the recognition and quantification of birds on WSR-88D products will enable meteorologists to determine when birds are biasing wind estimates. This information also will be of great importance to biologists who study bird movements through the atmosphere and those interested in flight safety and the avoidance of bird–aircraft collisions.

2. General methodology

The characteristics and details of operation of the WSR-88D can be found in Crum et al. (1993), Klazura and Imy (1993), and Crum and Alberty (1993). We began our work with the WSR-88D at the Houston, Texas, station (KHGX) in the spring of 1992. Houston was the first WSR-88D installed on the northern coast of the Gulf of Mexico. Initially we gathered data directly from the principal user processor (PUP) workstation by photographing with Fugichrome 100 film the displayed products with a 35-mm reflex camera mounted on a tripod. To prevent reflections on the screen from room light, we designed a hood (wooden frame with black felt) that could be placed easily in front of the screen whenever photography was allowed. The camera was set to automatic with an f-stop of 5.6 and a cable release was used to trigger the camera. We also followed these procedures at the WSR-88D site Slidell, Louisiana (KLIX) in spring of 1994. While at Slidell we were able to access other radars when they came on line (e.g., Mobile, Alabama; Lake Charles, Louisiana) as well as the Houston WSR-88D. We typically gathered images of the following products: base reflectivity (230-km range, 0.5° elevation), base velocity (same), VAD for different beam elevations, and VAD wind profiles.

In the spring of 1995 we evaluated the usefulness of a NEXRAD (Next Generation Weather Radar) information dissemination service and selected Unisys as a vendor. Images of base reflectivity (lowest elevation, 0.5°) and base velocity (same elevation) were collected by personal computer via modem once every hour during the spring (15 March–31 May) and the fall (1 September–30 November). These images have 1 km × 1 km resolution, 16 data levels, and 230-km range. Images were stored on optical disk for archiving and future analysis in the Radar Ornithology Laboratory at Clemson University. We also gathered additional base reflectivity images from different WSR-88D sites through the Internet at http://www.intellicast.com.

Direct visual studies of migration aloft were made during the spring and fall when possible in an effort to calibrate the reflectivity products of the WSR-88D. In the spring on the gulf coast, vertical observations with a 30 × telescope or 20 × 60 binoculars were made during the collection of radar data. When radar data were gathered at KHGX facility, daytime observations were made on Galveston Island, and when radar data were being collected at the Lake Charles, Louisiana station, daytime observations were made on the coast near the town of Cameron, Louisiana. For KLIX station, vertical daytime observations were made at the station. In the fall when only nocturnal movements of birds were studied, we used the technique of moon-watching (Lowery 1951) during periods when the moon was full and the sky was clear. By directing a 30 × telescope toward the disk of the moon, one could see the silhouettes of migrating birds, identify the general type of bird (e.g., songbird, waterfowl), and record directional coordinates that could be used to compute the number of birds crossing a 1.6-km line per hour (Lowery 1951). On nights without a full moon we used a vertically directed image intensifier (6.2 ×, second-generation crewserved weaponsight, AN/TVS-5, Varo) to monitor the flow of birds through a vertically pointing spotlight (300 W or J s−1) and followed the methods of Gauthreaux (1969) for making observations. The observer was aligned head north–feet south, and the flight directions of birds passing through the field of view were recorded as clock face coordinates (in at 10, out at 4), converted to azimuths, and subsequently analyzed with circular statistics (Fisher 1993).

3. Bird migration patterns on the WSR-88D

Although birds may be detected by weather radar at any time during the annual climatic cycle, they are most likely to be detected during the spring when they are migrating north to breed and during the fall when they are migrating south to reach areas where they spend the winter. Because of the energetic costs of transport, most birds use tailwinds for migration and are less likely to fly when winds are strong and opposing and when precipitation is heavy (see Richardson 1990, 1991). Weather permitting some migration occurs during almost every hour of the day during the spring and fall. Raptors, cranes, and other large-bodied soaring species migrate during daylight hours and use thermals to gain altitude and then glide between thermals. They follow a zigzag course of thermal drift in response to atmospheric motion and subsequent glide to compensate for the drift while in thermals (see diagram in Kerlinger 1989). When thermal development is limited by weather conditions, these species show little migratory behavior, because they must expend energy in powered (wing flapping) flight. Waterfowl, shorebirds, and many other species of water birds migrate during both daylight and darkness, but the greatest movements are initiated at the beginning of the evening. Most songbird species that migrate do so at night when the atmosphere is more conducive to powered flight (Kerlinger and Moore 1989) and seek shelter and feed during the daylight hours (Berthold 1996, 34). Overall the greatest amount of migration displayed on radar is typically during the first half of the night from 0130 UTC until 0530 UTC.

a. The arrival of transgulf bird migration in spring

The spring migration of land birds and shorebirds across the Gulf of Mexico begins in the first and second week of March, reaches a peak in late April and early May, and is essentially over by the third week in May. Only rarely do flights continue until the end of May (Gauthreaux 1971). The patterns of winds aloft over the Gulf of Mexico are critically important to the seasonal timing of transgulf migration. In March the winds over the gulf are influenced often by continental polar air masses (anticyclonic systems over the southeastern United States) and winds blow from the east near the surface and aloft. Only when southerly return flow of maritime tropical air occurs are conditions good for a south-to-north transgulf crossing, and this occurs aloft before it occurs on the surface. Consequently winds aloft are generally more favorable for a gulf crossing than are surface winds. As spring progresses the number of days with good return flow increases. In April and May the favorableness of the pattern of winds aloft increases as cold fronts decrease in frequency (see also Duncan 1994). When powerful cold fronts move southward over the gulf in late April and early May, spectacular “fallouts” of migrants often take place on offshore oil rigs and fishing boats in the northern gulf as birds are forced by hard rain and adverse winds to land wherever they can. The birds are often exhausted and so lean having catabolized breast muscle tissue after exhausting their fat supply that the keel of their sternum is like a knife blade. Many of these migrants do not survive. When cold fronts are weak and shallow, most transgulf migrants continue flying north in southerly airflow above the frontal boundary.

The arrival of a transgulf flight on the northern gulf coast is strongly influenced by weather conditions over the gulf (Gauthreaux 1971). Typically, with moderate southerly winds (5–8 m s−1, about 10–15 kt) the movements begin in the late morning hours, reach peak densities in the afternoon, and are largely finished by nightfall. The arrival time of a flight is strongly correlated with the velocity of the southerly winds aloft, and when wind conditions are favorable but low in velocity, a transgulf migration may not appear on the display until the afternoon and at nightfall migrants are still arriving from over the gulf. The display of the arriving transgulf migration on the WSR-88D is easily recognized and can be discriminated from precipitation-type echoes in base reflectivity products (Figs. 1a–f). The first indication on the base reflectivity product (lowest elevation) of the arriving transgulf flight occurs at 1610 UTC on the KLCH WSR-88D as small, scattered echoes appear offshore (Fig. 1a). As the leading edge of the movement approaches the coast the density of the movement increases and reflectivity values of pulse volumes increase from 5–20 dBZ (Figs. 1d–f). In Figure 1e some of the pulse volumes containing birds have reflectivities of 20 dBZ. Most of these values are positioned on the display west and east of the radar where the flying birds present the maximum cross-sectional area to the radar beam (e.g., the body axis oriented south–north and perpendicular to the radar beam). In favorable flying conditions most of the transgulf migrants fly over the coast and make landfall in the bottomland forests and woodlands from 50 to 120 km inland from the coast. Only during adverse weather (hard rain or strong north winds) do substantial numbers put down in coastal woodlands.

b. The nocturnal exodus of migratory birds

Thirty to 40 min after sunset (during nautical twilight), most of the migrants that landed earlier in the day depart the stopover areas and continue their migratory journey. The exodus is often spectacular on the WSR-88D as the density of migrants aloft increases rapidly. In Figs. 2a–e the exodus of migrants from stopover areas begins near 0116 UTC and overlaps the continued arrival of migrants from over the gulf. The maximum values of reflectivity quickly reach 25 dBZ (Figs. 2a and 2b). Four hours after the beginning of the exodus (0520 UTC) (Fig. 2c), the number of pixels showing 20 dBZ has decreased, and by 0716 UTC (Fig. 2d) no pulse volumes have reflectivities above 15 dBZ. By 0912 UTC the migration from over the gulf has finished and the echoes that remain are from smaller numbers of migrants moving toward the NE from the Texas coast (Figs. 2e and 2f). The large precipitation echo to the north of the surveillance area can be clearly distinguished from the echoes produced by birds. The duration of the exodus rarely exceeds 3 h for radar stations on the central northern gulf coast, but for stations on the Texas coast, the movement of migrants toward the NE continues for most of the night. These movements contain birds that arrived from over the gulf as well as migrants moving up the Texas coast from eastern Mexico (Forsyth and James 1971).

Because the WSR-88D is a Doppler radar, it provides information on the direction of target movement as well as information on radial velocities in the base velocity product. Although the differences in radial velocities are difficult to discern from the colored images of Fig. 3, the majority of the echoes moving directly toward the radar station have velocities of 18.6–30.6 m s−1. Because these targets are approaching the radar site, the velocities are true ground speeds. When targets are moving perpendicular to the radar beam, their radial velocities are zero. In Fig. 3 the gray line denotes zero radial velocity and the flow of migrants is perpendicular to this line. On the northern gulf coast in spring, when migrant birds arrive from over the gulf and land just inland from the coast, the “inflow” velocities south of the radar stations greatly exceed the “outflow” velocities (Fig. 3a). After dark, when the migrants that landed earlier in the day take off and resume their migration, the outflow velocities typically exceed inflow velocities if no birds continue to arrive from over the gulf. In Fig. 3b this is not the case as some birds continue to arrive after dark while large numbers are departing stopover areas and moving toward the NNE. In this case the inflow and outflow values are comparable.

c. The effects of bird migration on meteorological data

How do the displays of bird migration on the base reflectivity and base velocity products of the WSR-88D relate to weather conditions at the time? Figure 4 shows the surface map and the 850-mb winds aloft map on 30 April 1995 at 1200 UTC. Prior to the arrival of the transgulf flight on 30 April 1200 UTC, the surface map shows essentially calm conditions at Lake Charles (200° at 2 m s−1 from radiosonde data), but the 850-mb winds aloft map shows winds from 240° at 10.7 m s−1 (from radiosonde data) over the station. At 0000 UTC when transgulf migration was arriving on the northern gulf coast and passing over the Lake Charles National Weather Service (NWS) station, the radiosonde data show surface winds at Lake Charles from 185° at 3.4 m s−1, and the winds at 850 mb from 205° at 8.2 m s−1 (Fig. 5). These velocities are in stark contrast to the VAD wind profile data from the WSR-88D at Lake Charles for the same time period (Fig. 6). The VAD wind profile data for each of the volume coverage patterns between 0001 and 0059 UTC indicate winds of 20.7 m s−1 for the altitudes between 305 and 3049 m. It is clear that migrating birds are biasing the winds aloft information generated by the WSR-88D VAD algorithms. The birds so influence the VAD wind profiles that they often can be used to document the altitudinal distribution of the migration (see Haro and Gauthreaux 1997; Gauthreaux et al. 1998). Not only do migrating birds bias the VAD wind profile data, but their appearance on the WSR-88D base reflectivity products can apparently bias the distribution of precipitation mapped in surface weather maps. In Fig. 5 precipitation is indicated near Houston and Lake Charles weather stations, but an examination of the base reflectivity product from KLCH for the same time period (Fig. 1f) shows only a pattern of bird echoes with some pulse volumes showing 20-dBZ reflectivity. The Houston WSR-88D at 0000 UTC 1 May 1995 was in clear air mode and showed arriving transgulf migration with maximum reflectivity values of 28 dBZ 74–130 km to the east of the station.

4. Bird roosting movements on the WSR-88D

Outside of the spring and fall migration seasons the WSR-88D readily detects concentrations of birds (and bats) as they depart and return to roost sites. The departures of birds from roost sites at dawn and near sunrise are particularly prominent on base reflectivity products as the radar beam is often bent back toward the ground because of superrefraction. Inversions of temperature and moisture influence the index of refraction of the radar beam in the atmosphere, and these conditions are characteristic of the lower atmosphere near dawn when the birds depart their overnight roost (Russell and Gauthreaux 1998). In contrast, near the time of sunset and dusk normal propagation of the radar beam usually occurs, and the beam is often too high to detect the bird targets returning to the roost unless they are flying at altitudes covered by the radar beam. Figure 7 shows the distribution of several purple martin (Progne subis) roosts as detected by the Fort Polk, Louisiana, WSR-88D on the mornings of 8 and 9 July 1995. In the upper image of Fig. 7 the radar is in clear air mode and roosts are located to the NNW, NE, and SW of the station. The SW roost site is near Beaumont, Texas. In the lower image of Fig. 7 two additional roost sites are indicated: one to the SE near Baton Rouge, Louisiana, and another to the WSW in Texas. The likely reason more ring echoes show up on the second morning is because the refractive index of the atmosphere is bending the radar beam downward to a greater extent. Although not illustrated, the martins in these images typically fly at 15–18 m s−1 (Russell and Gauthreaux 1998). As the annulus of martins on the display grows in diameter the reflectivities decrease, and as the superrefraction conditions disappear the radar beam bends and overshoots the birds. During the winter months, large roosts of “blackbirds” [e.g., icterids and starlings (Sturnus vulgaris)] are often detected by the WSR-88D (Gauthreaux, personal observation), but no systematic analyses of these data have been made.

5. The quantification of nocturnal bird migration on the WSR-88D

During the fall of 1993 quantification of the density of nocturnal bird migration displayed on the WSR-88D began. The amount of migration passing over the radar station was measured using the technique of moon-watching (Lowery 1951; Lowery and Newman 1955). Moon-watching data were gathered within two nights of the full moon when the elevation of the moon was above 30° and the weather conditions were acceptable for moon-watching (nearly cloudless) and for bird migration (favorable winds for the season and grounded birds ready to migrate). A total of 50 moon-watch samples were taken on 21 dates between 1 October 1993 and 9 October 1995 near Houston, Lake Charles and Slidell, and Greenville, South Carolina. The moon-watch data were analyzed according to the methods discussed in Nisbet (1959, 1961, 1963a,b).

Photographs of base reflectivity products (lowest antenna elevation of 0.5°) on the PUP were taken to correspond with each moon-watching sample. Each photograph was scored such that the highest decibel value of reflectivity in the display of bird echoes was coded. For example, in Fig. 2b the maximum value is 25 dBZ. This base reflectivity product shows dense migration leaving southwestern Louisiana. In Fig. 2d, the maximum value is 15 dBZ.

Figure 8 shows the relationship between the maximum pixel value measured in decibels of reflectivity on the base reflectivity products and the corresponding measure of migration traffic rate [number of birds crossing a 1.6 km (1 mi) line h−1.] The best fit line is a third-order polynomial and the relationship is highly significant (r2 = 0.87; F ratio = 104.83, P < 0.00001). With this relationship one can estimate the amount of bird migration within 100 km of a WSR-88D anywhere in the United States. All of the WSR-88D radars adhere to strict calibration standards to ensure that they see the same precipitation at the same intensity. Because dense swarms of insects in the atmosphere produce base reflectivity displays that are quite similar to low-density bird migration displays (see Russell and Wilson 1996), it is essential that base velocity information be gathered at the same time as base reflectivity products. With a knowledge of the winds aloft it is possible to distinguish bird movements from insect movements, because the latter have air speeds that rarely, if ever, exceed 8–10 m s−1 (most are between 4 and 6 m s−1) while migrating birds typically fly at speeds greater than 10 m s−1. Because of their low air speeds insects tend to move with the wind, deviating just a few degrees from down wind, whereas birds will sometimes fly against the wind. In general, insect ground speeds exceed corresponding wind speeds by approximately 2–6 m s−1 (Schaefer 1976).

6. Conclusions

With the development of the new national network of WSR-88D (NEXRAD) weather surveillance radars in the United States, it is now possible to use the units for biological studies of the atmosphere. Because of the sensitivity, power, and resolution of the WSR-88D, it will play an important role in the detection, quantification, and monitoring of bird flights and those of bats and insects. These new radars can readily detect birds in the atmosphere and can play an important role in warning pilots of hazardous concentrations of birds thus reducing the likelihood of serious bird–aircraft collisions. The need for recognizing biological targets on the displays of the WSR-88D cannot be overstated. Such targets, particularly migrating birds, seriously bias information on wind speeds measured by the radar and also likely bias algorithms that automatically map precipitation echoes on national mosaic displays of WSR-88D data.

The recognition of migrating birds on weather surveillance radar will also clarify many of the misconceptions that television weathermen have about “noisy” or “anomalous” radar patterns that appear on national mosaics. These mosaics usually display reflectivity levels of 20 dBZ and above, because the lower reflectivity levels have been eliminated or filtered out. Despite setting the minimum base reflectivity filters at 20 dBZ, during dense bird migrations, reflectivity values from the mass of migratory birds may reach 20 dBZ levels and even reach 30-dBZ level during the seasonal peak of migration. Such patterns are commonplace at night during the spring and fall migration periods and it would be informative to the viewing public if television weathermen pointed out that such displays are caused by dense concentrations of migrating birds and are not attributable to meteorological targets.

Acknowledgments

Our work with the WSR-88D would not have been possible without funding from the Department of Defense, Legacy Resource Management Program, and the cooperation of the radar and forecasting personnel of the National Weather Service WSR-88D stations at Houston (KHGX)—Bill Read (meteorologist in charge); Lake Charles (KLCH)—Dave McIntosh (meteorologist in charge); and Slidell (KLIX)—Bill Crouch (former meteorologist in charge) and Paul Trotter (current meteorologist in charge). Additional support came from the Houston Audubon Society, the National Fish and Wildlife Foundation, the U.S. Fish and Wildlife Service, and the National Biological Survey at the Southern Science Center, Lafayette, Louisiana. We thank David S. Mizrahi for valuable assistance with data analysis and preparation of the manuscript, and Rob Steenburgh and two other reviewers for their careful and helpful reviews of the manuscript.

REFERENCES

  • Able, K. P., 1972: Fall migration in coastal Louisiana and the evolution of migration patterns in the gulf region. Wilson Bull.,84, 231–243.

  • Berthold, P., 1996: Control of Bird Migration. Chapman and Hall, 355 pp.

  • Crum, T. D., and R. L. Alberty, 1993: The WSR-88D and the WSR-88D operational support facility. Bull. Amer. Meteor. Soc.,74, 1669–1687.

    • Crossref
    • Export Citation
  • ——, ——, and D. W. Burgess, 1993: Recording, archiving, and using WSR-88D data. Bull. Amer. Meteor. Soc.,74, 645–653.

    • Crossref
    • Export Citation
  • Duncan, R. A., 1994: Bird migration, weather and fallout: Including the migrant traps of Alabama and northwest Florida. 95 pp. [Available from R. A. Duncan, 614 Fairpoint Dr., Gulf Breeze, FL 32561.].

  • Eastwood, E., 1967: Radar Ornithology. Methuen, 278 pp.

  • Fisher, N. I., 1993: Statistical Analysis of Circular Data. Cambridge University Press, 277 pp.

    • Crossref
    • Export Citation
  • Forsyth, B. J., and D. James, 1971: Springtime movements of transient nocturnally migrating landbirds in the gulf coastal bend region of Texas. Condor,73, 193–207.

    • Crossref
    • Export Citation
  • Gauthreaux, S. A., Jr., 1969: A portable ceilometer technique for studying low-level nocturnal migration. Bird-Banding,40, 309–320.

    • Crossref
    • Export Citation
  • ——, 1970: Weather radar quantification of bird migration. BioScience,20, 17–20.

    • Crossref
    • Export Citation
  • ——, 1971: A radar and direct visual study of passerine spring migration in southern Louisiana. Auk,88, 343–365.

    • Crossref
    • Export Citation
  • ——, 1972: Behavioral responses of migrating birds to daylight and darkness: A radar and direct visual study. Wilson Bull.,84, 136–148.

  • ——, 1992: The use of weather radar to monitor long-term patterns of trans-Gulf migration in spring. Ecology and Conservation of Neotropical Migrant Landbirds, J. M. Hagan III and D. W. Johnston, Eds., Smithsonian Institution Press, 96–100.

  • ——, D. S. Mizrahi, and C. G. Belser, 1998: Bird migration and bias of WSR-88D wind estimates. Wea. Forecasting,13, 465–481.

    • Crossref
    • Export Citation
  • Haro, J., and S. Gauthreaux, 1997: WSR-88D VAD wind profile data influenced by bird migration over the southwest United States. NOAA Tech. Memo. NWS WR-244, 26 pp. [Available from National Technical Information Service, U.S. Dept. of Commerce, 5285 Port Royal Rd., Springfield, VA 22161.].

  • Kerlinger, P., 1989: Flight Strategies of Migrating Hawks. University of Chicago Press, 375 pp.

  • ——, and F. R. Moore, 1989: Atmospheric structure and avian migration. Current Ornithology, D. Powers, Ed., Vol. 6, Plenum Press, 109–142.

    • Crossref
    • Export Citation
  • Klazura, G. E., and S. A. Imy, 1993: A description of the initial set of analysis products available from the NEXRAD WSR-88D system. Bull. Amer. Meteor. Soc.,74, 1293–1311.

    • Crossref
    • Export Citation
  • Lack, D., and G. C. Varley, 1945: Detection of birds by radar. Nature,156, 446.

    • Crossref
    • Export Citation
  • Larkin, R., 1984: The potential of the nexrad radar system for warning of bird hazards. Proc. Wildlife Hazards to Aircraft Conf. and Workshop, Charleston, SC, FAA, 369–379. [Available from Dept. of Transportation, Federal Aviation Administration, Office of Airport Standards, Washington, DC 20591.].

  • Lowery, G. H., Jr., 1951: A quantitative study of the nocturnal migration of birds. Univ. Kans. Publ. Mus. Nat. Hist.,3, 361–472.

  • ——, and R. J. Newman, 1955: Direct studies of nocturnal bird migration. Recent Studies in Avian Biology, A. Wolfson, Ed., University of Illinois Press, 238–263.

  • Nisbet, I. C. T., 1959: Calculation of flight directions of birds observed crossing the face of the moon. Wilson Bull.,71, 237–243.

  • ——, 1961: Studying migration by moon-watching. Bird Migr.,2, 38–42.

  • ——, 1963a: Estudio de la migracion sobre el disco lunar (I). Ardeola,8, 5–17.

  • ——, 1963b: Quantitative study of migration with 23-centimetre radar. Ibis,105, 435–460.

    • Crossref
    • Export Citation
  • Richardson, W. J., 1990: Timing of bird migration in relation to weather: Updated review. Bird Migration: Physiology and Ecophysiology, E. Gwinner, Ed., Springer, 78–101.

    • Crossref
    • Export Citation
  • ——, 1991: Wind and orientation of migrating birds: A review. Orientation in Birds, P. Berthold, Ed., Birkhäuser Verlag, 226–249.

  • Russell, K. R., and S. A. Gauthreaux Jr., 1998: Use of weather radar to characterize movements of roosting purple martins. Wildl. Soc. Bull.,26 (1).

  • Russell, R. W., and J. W. Wilson, 1996: Aerial plankton detected by radar. Nature,381, 200–201.

    • Crossref
    • Export Citation
  • Schaefer, G. W., 1976: Radar observations of insect flight. Insect Flight, R. C. Rainey, Ed., Blackwell, 157–197.

  • Williams, T. C., J. M. Williams, L. C. Ireland, and J. M. Teal, 1977:Autumnal bird migration over the western North Atlantic Ocean. Amer. Birds,31, 251–267.

Fig. 1.
Fig. 1.

WSR-88D base reflectivity images (0.5° elevation) depicting a transgulf migration of birds arriving at Lake Charles, LA, 30 April/1 May 1995.

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 2.
Fig. 2.

WSR-88D base reflectivity images (0.5° elevation) depicting the departure of nocturnal bird migrants from stopover sites around Lake Charles, LA on 1 May 1995. Note the large precipitation echo moving into the surveillance area to the north of the station (c)–(f).

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 3.
Fig. 3.

WSR-88D base velocity images (0.5° elevation) depicting a transgulf migration of birds arriving at Lake Charles (a) on 30 April 1995 and (b) the departure of nocturnal bird migrants from stopover sites around Lake Charles on 1 May 1995.

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 4.
Fig. 4.

(a) Surface conditions and (b) 850-mb winds aloft on 30 April 1995.

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 5.
Fig. 5.

(a) Surface conditions and (b) 850-mb winds aloft on 1 May 1995.

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 6.
Fig. 6.

WSR-88D VAD depicting wind profiles at Lake Charles 0001-0059 UTC, 1 May 1995.

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 7.
Fig. 7.

WSR-88D base reflectivity images (0.5° elevation) depicting annular signatures associated with morning departures of purple martins from roost sites. Images show ring echoes in both clear air mode (upper panel, 8 July 1995) and precipitation mode (lower panel, 9 July 1995).

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

Fig. 8.
Fig. 8.

Relationship between number of birds crossing a 1.6 km line h−1 (migration traffic rate) and the maximum reflectivity (dBZ) detected by the WSR-88D.

Citation: Weather and Forecasting 13, 2; 10.1175/1520-0434(1998)013<0453:DOBMOT>2.0.CO;2

* An electronic supplement to this article may be found on the CD-ROM accompanying this issue or at http://www.ametsoc.org/AMS.

Save