Receivers. At the beginning we used AVM LA 12-DS receivers (AVM Instrument Co.). These are comparatively heavy, but well protected receivers, and are specially suited for car tracking. We also tested WMI TR X 12 (Wildlife Materials, Inc.). Both are powered by 8 alkaline batteries (1.5V each)
Afterwards we adopted Telonics receivers, equiped with a scanner which make it easier to check continuously many birds, when the quantity of transmitters is relatively high. They can be powered by 12 NiCd rechargeable C-cells (1.2V each) or standard alkaline batteries (1.5V each). Telonics receivers are very expensive, but are probably the best at the moment. 

Antennas. We have employed both directional and omnidirectional antennas. Directional antennas are 3-element hand-held yagi, and can be either standard or folding. They are so called because they give a better beep signal when the antenna is pointing in the direction of the transmitter. We used a 2-element directional antenna manufactured by Telonics (see pictures down this page). Omnidirectional antennas have a lower receiving range, but are suited for tracking huge areas by car, for instance. We also use an omnidirectional antenna while tracking birds by aircraft (see aircraft procedures).

Harnesses. The transmitters were attached to the birds following procedures detailed by Kenward (1979). In the beginning we used harnesses made of plastic coated wire, and consisting of two loops that cross each other under the breast, the crossing point being fixed with two component epoxi to prevent shifting of the harness. On a second step we substituted the plastic coat by the more durable surgical tubing, to cover the wire. Finally the harnesses were made a Teflon, much more expensive but softer and enough durable to keep the transmitters on the birds as long as batteries would work. Teflon has the advantage that the transmitter can get lost after the batteries are exhausted.

	Picture by J.A. Alonso
Picture by M.I. Madrid

Heigh is a crucial factor, both for the transmitter or the receiver. And this is not always enough stressed  by the manufacturer. Sometimes a signal is dificult to recognize from the ground but is perfectly audible from the top of a car. Ground to ground average distance could be 1-5km, depending on ground relief. If there is some high spot -a small hill, a tower, any high building- the receiving distance can increase up to 10km. Flying birds signals can be received at very long distances, 10-20km, while an aircraft allows the observer to get signals of landed birds up to 40-60km. When on aircraft, a flying bird can be located at distances that could reach 80-100km.

AERIAL TRACKING IN SPAIN AND GERMANY

Introduction

Aerial tracking was first used in Spain in 1988 to locate Spanish Imperial Eagle (Aquila adalberti) young after their independence in Doñana. Two of the authors of the present paper continued to use this method to locate wintering Common Cranes (Grus grus) in Extremadura after marking them during their migratory staging at Gallocanta, Zaragoza province, following mainly the procedures described for aerial tracking of Whooping (G. americana) and Sandhill (G. canadensis) Cranes (Brander & Cochran, 1971; Cochran, 1972, 1980; Drewien & Bizeau, 1981; Gilmer et al., 1981; Melvin & Temple, 1982). From 1992 on, aerial tracking has also been used to study dispersal behaviour of radiomarked Great Bustards (Otis tarda). In Germany, aerial tracking started in June 2000 to search for the immature cranes comming back to their natal areas after family break-up during first winter or after spring migration (see also Dispersal of juvenile and subadult German birds: what do they do during their immature period?).

The combined work of a ground team and an airborne tracking crew is necessary to recover contact with individuals that move out of land radioreceiver reach. Such recovery is specially important when continuous behavioural observations and data on social structure or habitat preferences are required.

Initially, we used Dornier aircraft, given their plane upper position and low minimal speed, what allows visualization of tracked birds. Sighting of birds from aircraft is highly useful in Cranes, but practically impossible in Great Bustards due to their tendency to fly when they detect aircraft approach. Great Bustard behaviour requires greater flight altitude to prevent the birds' take off by disturbance, which may cause the ground team not to find the individuals tracked. Since 1992, most birds were located from Bonanza E-24 airplanes, much faster than Dorniers and which allow to reach easily higher altitudes.

Materials and Methods

Telemetry equipment. The equipment used for aerial tracking is the same one used in land work, that is, a TELONICS receiver (TR-2 model, provided with a TS-1 model scanner), supplied either with battery, either alcaline or NiCd, power or with power from the plane's electric system. The transmitters attached to Great Bustards were different models from trademarks TELONICS, BIOTRACK and TELEVILT.

Aerial tracking was started for Common Cranes following the procedures previously applied to eagle location. We flew "high-wing" aircraft, Dornier E-9, equiped with WMI three element Yagi direccional antennas to each wing strut, mounted in a similar way to that described by Gilmer et al. (1981). This allowed us to put both antennas in a slightly side-looking configuration (about 30 degrees from flight direction). Antenna elements were always vertically polarized. And this is also the way we now track immature birds when they come back in Germany from the wintering areas, although the aircraft used there is a Cessna 172 Skyhawk.

General procedures. All flights are VFR, although flight altitude is always high enough to prevent birds to move long distances because of aircraft flying over. When searching for recently lost individuals we start aerial tracking over the spot of last ground visual contact with the bird. If no signal is detected during the approach to that point, a 2 km radius circular flight is usually carried out around it, at an altitud of 500-800 m above ground level, while scanning through the directional antenna. In most cases, signals are received while approach the last visual contact coordenate, at a longest distance of about 20-30 km, depending on flight high. Detection distance depend on transmitter power and bird position and activity in that moment. Reported maximum reception distances refer to birds tracked during migration, with a headstart up to 160 km -4 hours flying- on a known route (Melvin, 1982).

As usually happens with directional antennas, signal quality is optimal when the antenna is pointing directly to the transmitter, and it almost disappears when pointing in a right angle side direction. Once the right direction is fixed, we fly straight towards the signal lowering flight altitude (about 120m a.g.l.) to increase location accuracy. The directional antenna is then selected. Since directional antenna elements are perpendicular to ground, when the aircraft pass over the exact point in which a transmitter is located signal intensity suddenly fall, thus indicating transmitter position. The exact point is then visually spotted and, at the same time, the signal is switched to the omnidirec­tional antenna. Signal reception quality has to be then as good as it was when reception was switched to the omnidirectional antenna, just before flying over the exact transmitter location. To assure correct location, the plane is usually turned around 180 degrees to fly back over the same spot and determine its coordinates by means of a GPS. If signal intensity has not clearly fallen, a better location has then to be achieved. To do so the airplane is turned to describe a 3 or 4 km curve and spot the exact location in a direction perpendicular to the previous one. If this second localization is correct, its coordenates are determined by GPS and written down for land control.

When surveying large areas in search of long lost radiomarked birds, flight altitude is 350-450m a.g.l., scanning through both directional and omnidirectional antennas. Transect width at that altitude is between 10 and 15 km to either side of the plane, slightly less than the maximum band width of 24 km to either side reported for two side-looking antennas on high-wing aircraft (Melvin and Temple, 1987). This flight altitude is a good compromise between maximizing reception range and minimizing time required for descents to accurately locate radiomar­ked birds and subsequently climb again to resume searching. Some authors report maximum signal gain not being increased at altitudes greater than 300 m a.g.l. over open areas (Gilmer et al., 1981), although higher flight altitudes might increase sensityvity over mountains and forest areas. Other authors indeed found an increase reception range above 300 m, which they relate to the low position of the transmitting antenna on leg band mounted radios (Melvin and Temple, 1987). Nevertheless, we assume that our flight altitude is high enough to receive any signal in the mentioned range, even in the case of birds lying on the ground or in slightly hilly habitats.


Results

Benefits and handicaps of aerial tracking compared to ground tracking. Aerial tracking is used only when some indvidual could no longer be contacted with by conventional ground tracking, given the great importance of continuous contact with marked individuals to our study. High location altitude in the case of Common Cranes prevents bird visualization, for subsequent ground location is needed to continue observations. Aerial radio-monitoring is not always possible due to unfavoralbe weather conditions which caused the delay of 75% of flights.

Reception range from ground is between 10 to 20 times smaller than from aircraft. Frequently in our studies, ground crew mobility is limited by the availabilty of tracks and roads when birds visited new areas unknown to observers. Land crews may be, therefore, realtively inefficient to radiotrack birds in a large and poorly known areas.

Tracking Common Cranes from aircraft gives the highest probabilities to locate any lost radiomarked bird. Besides, aerial location allow us to study the general topography of routes followed by Common Cranes in their movements. Mean distance between aerial coordenates and subsequent ground locations was only 576 m.

Quite frequently we had to lower our long-distance tracking flight altitude due to heavy interference from powerful radio sources, such as big towns, airports, etc. The only way to avoid such interferences is to change flight direction so that the plane is between the source of interference and the area to be monitored, that is, the interference source behind the directional antenna. These maneuvers increase searching time and reduce the widht of transect band.


Reasons for losing the signal. In relation to factors that may cause the loss of radio signals, the most frequent ones are transmitter type and power, as well as transmitter life. Transmitters are much more difficult to locate during the last phases of their life. This difficulty is lowered when birds are tracked from the air. Type of transmitter attachment may also affect signal quality, for the closest the antenna is to the bird's body (backpacks) or to the ground (leg-mount), the worse reception gets. Antenna position is also very important for reception quality and depends mainly on the bird's activity (flying, feeding or vigilant, lying on the ground, dead upside down, dead inside a fox den; transmitter thrown on the ground or buried, etc.). Geographical accidents such as hilly terrain, narrow valleys, river banks or canals,  may also set a difficulty on correct localization, although it is not usually the case in the Common Crane.


Routine in the case of not receiving a signal. When ground contact with a given transmitter is lost, it must be searched for from the air, beginning at the coordinates where it was received last time. If the bird's range previous to signal loss showed any directionality, tracking must be carried out in those areas ahead the bird's historical movements. The greater flight altidude, the greater the probability to reestablish radio contact and the longer the signal recovery distance.

            After locating a new frequency in the scanner, signal strength may fluctuate as a consecuence of the bird's movements, often resulting in false nules and peaks. This is specially evident when birds are still far away from aircraft. Flight direction should then be changed; i.e. input through the directional antenna, until the right direction can be fixed and an accurate bearing to a true peak signal is determined.


Acknowledgements

Aerial radiotracking of birds in Spain would have never been possible without the disinterested colaboration of the Spanish Air Forces and the staff of the Getafe Air Base. We wish to thank Group 42, as well as the many volunteer pilots who have piloted aircraft during all these years, and also the ground staff taking care of the tracking materials installed on plains.


Literature

ALONSO, J.A., E. MARTÍN, M.B. MORALES & J.C. ALONSO. 1996. Aerial tracking of Great Bustards (Otis tarda) in Spain. Int. Simposium for the Conservation of Steppe Birds and their Habitat. Valladolid 1995.

BRANDER, R.B. & W.W. COCHRAN. 1971. Radio-location telemetry. Pp. 95-104 in Giles, R.H. Jr. ed. Wildlife Management Techniques. The Wildlife Soc. Washington, D.C.

COCHRAN, W.W. 1972. Long-distance tracking of birds. Pp. 39-59 in Galler, S.R. et al., eds. Animal Orientation and Navigation. NASA SP-262.

COCHRAN, W.W. 1980. Wildlife Telemetry. Pp. 507-520 in Schemnitz, S.D. ed. Wildlife Management Techniques Manual. The Wildlife Soc. Washington, D.C.

DREWIEN, R.C. & E.G. BIZEAU. 1981. Use of radiotelemetry to study movements of juvenile whooping cranes. Pp. 130-134 in Lewis, J.C. & H. Masatomi, eds. Crane Research Around the World. Intl. Crane Found., Baraboo, Wisconsin.

GILMER, D.S., L.M. COWARDIN, R.L. DUVAL, L.M. MECHLIN, C.W. SCHAIFFER & V.B. KUECHLE. 1981.Procedures for the use of aircraft in wildlife biotelemetry studies. U.S. Dept. Interior, Fish and Wildlife Serv. Res. Publ. 140. 19 pp.

MELVIN, S.M. 1982. Migration ecology and wintering grounds of sandhill cranes from the Interlake region of Manitoba. Ph.D. Thesis. Univ. of Wisconsin, Madison. 263 pp.

MELVIN, S.M. & S.A. TEMPLE. 1982. Migration ecology of sandhill cranes: a review. Pp. 73-87 in Lewis. J.C. ed. Proc. 1981 Crane Workshop. Natl. Audubon Soc., Tavernier, Florida.

MELVIN, S.M. & S.A. TEMPLE. 1987. Radio telemetry techniques for international crane studies. Pp. 481-492 in Archibald, G.W. & R.F. Pasquier, eds. Proc. 1983 Intl. Crane Workshop. Intl. Crane Found., Baraboo, Wisconsin.