Effects of Radio Packages on Wild Ducks Author(s): David S. Gilmer, I. J. Ball, Lewis M. Cowardin, John H. Riechmann Reviewed work(s): Source: The Journal of Wildlife Management, Vol. 38, No. 2 (Apr., 1974), pp. 243-252 Published by: Allen Press Stable URL: http://www.jstor.org/stable/3800730 . Accessed: 06/02/2012 11:55

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management.

http://www.jstor.org EFFECTSOF RADIO PACKAGESON WILD DUCKS

DAVID S. GILMER,James Ford Bell Museum of Natural History, University of Minnesota, Minneapolis 554551 I. J. BALL,James Ford Bell Museum of Natural History, University of Minnesota, Minneapolis 55455 LEWISM. COWARDIN, U.S. Bureau of Sport Fisheries and Wildlife, Northern Prairie Wildlife Research Center, James- town, North Dakota 58401 JOHN H. RIECHMANN,James Ford Bell Museum of Natural History, University of Minnesota, Minneapolis 55455

Abstract: A total of 211 wild, free-flyingmallards (Anas platyrhynchos)and wood ducks (Aix sponsa) were equippedwith breast-mountedradio packagesduring the breedingseasons of 1968-72. Known predationloss was 7.6 and 12.0 percentfor mallardsand wood ducks,respectively; 60 percentoccurred within3 weeks of instrumentation.The highestpredation rate for mallardswas 0.0048kills per tracking day and 0.0136for wood ducks. A higherdirect recovery rate for instrumentedbirds (19.5 percent)than noninstrumentedbirds (8.1 percent) was probablydue to the novelty of the transmitterto hunters. Departurepatterns and locationsof directrecoveries were similarbetween radio-equippedand normal- bandedbirds. Amongfemale wood duckswith radios,recovery rates were lower than expected.Hunters indicatedthat 84 percentof the instrumentedducks recovered were in good or excellentcondition. Re- capturesof ducksas long as 1 yearafter being equippedwith radiopackages indicated that featherwear and skin abrasionwere not serious.A high rate for feedingon land by instrumentedmallards was prob- ably due to our abilityto moreeasily locate and observethese birdsin cover. Preeningrates were higher for instrumentedducks. As the birds becameadjusted to the package,preening decreased and feeding on waterincreased. Social and breedingbehavior of instrumentedducks did not appearto be adversely affected by the radio package. J. WILDL. MANAGE. 38(2):243-252

Radiotracking techniques are widely used dispersal, or growth rates of marked birds. in wildlife studies (Schladweiler and Ball Boag (1972) evaluated effects of radio 1968, Will and Patric 1972); however, the packages on the behavior of 13 male and effects of radio packages (transmitter, bat- 7 female pen-reared captive red grouse tery, and harness) on behavior and survival (Lagopus 1. scoticus). He reported that of marked animals are poorly understood. food consumption of females was signifi- It is difficult to make comparisons with cantly reduced during the first 6 days after controls because the same factors that dic- attachment of radio packages and that ac- tate the use of telemetry often preclude tivity levels were reduced in both sexes for gathering data on unmarked animals. approximately 2 weeks. Habitat choice did Several studies have been undertaken to not differ between radio-equipped and un- evaluate the effects of radio packages on marked grouse. Greenwood and Sargeant the and condition behavior physical of (1973) measured the effect of back-mounted instrumented birds. Johnson (1971) leg- radio packages on the physical condition banded and released 785, 11-week-old male of 60 pen-reared captive mallards and blue- ring-necked pheasants (Phasianus colchi- winged teals (Anas discors). Birds equipped cus); 393 were equipped with radio- with radio packages lost more weight than packages in which transmitter and battery and skin irrita- were simulated. He concluded from re- did controls. Feather wear tion occurred on all marked individ- ported recoveries and bag checks that the nearly that radio packages did not affect the survival, uals. Periodic observations suggested treated birds preened more than control 'Present address: U.S. Bureau of Sport Fisheries birds and exhibited a partial aversion to and Wildlife, Northern Prairie Wildlife Research Center, Jamestown, North Dakota 58401. swimming.

J. Wildl. Manage. 38 (2):1974 243 244 EFFECTS OF RADIOS ON WILD DUCKS * Gilmer et al.

Table 1. Adult mallards and wood ducks radio-equipped on the Chippewa National Forest study area in north-central Minnesota, 1968-72.

Year Species Sex 1968 1969 1970 1971 1972 Total Mallard F 12 14 26 20 13 85 M 5 3 6 10 9 33 Wood duck F 10 20 15 17 0 62 M 3 6 10 12 0 31 Total 30 43 57 59 22 211

Studies cited above were designed to test Natural History, University of Minnesota, the effect of transmitters on the survival Minneapolis. and behavior of instrumented birds and, to reduce variation, restricted attention to METHODS birds or birds released captive pen-reared the field season 1968 into the wild. Information in this During through reported 211 adult mallards and wood ducks was derived from a of the 1972, paper study were with radio of ducks. we equipped packages (Table breeding ecology Therefore, Other birds instrumented but not con- did not to control variables such 1). attempt sidered in this included 13 ducks as duration of instru- analysis capture techniques, tracked the late summer of 1967 in or the and of during mentation, length frequency order to test the of the te- visual observations on marked and un- performance and the harness marked birds. these the lemetry equipment design. Despite drawbacks, 64 of the adult mal- of wild ducks with Approximately percent large sample free-flying lards and wood ducks were instrumented operating transmitters furnished informa- and the remainder, dur- tion that can add to our of during April May; understanding and A total of 61 the effect of radio on wild ducks. ing June, July, August. packages adult mallards and 161 adult wood ducks This took a study place during 5-year period were banded but not instrumented between in the National Forest of north- Chippewa 1968 and 1972. Most of these birds were central Minnesota, 19 km east of Bemidji. captured in July and August. Ducks were Project funds were provided by the U.S. captured using a variety of methods at Bureau of Sport Fisheries and Wildlife; by various locations within a 98 km2 study area. National Institute of Health Training Grant The region is forested and has a wide va- 2T01 GM 01779-06, and by the U.S. Atomic riety of lake, marsh, and river habitats. The Commission Energy (C00-1332-84). radio package design (Ball 1971:6) con- We acknowledge the assistance of J. R. sisted of a breast-mounted transmitter and D. F. D. H. A. Tester, McKinney, Johnson, battery attached by means of a body-loop B. Sargeant, R. J. Greenwood, and P. F. antenna and a neck loop. A small name tag Springer in reviewing the manuscript. D. A. providing the address of the investigators Davenport assisted with computer process- was fastened to transmitters used after 1968. ing. Radiotracking equipment used in this Average package weights during 1968 were study was supplied by the Bioelectronics about 29 and 27 grams for mallards and Laboratory, James Ford Bell Museum of wood ducks, respectively; but after 1968

J. Wildl. Manage. 38 (2):1974 EFFECTSOF RADIOSON WILD DUCKS* Gilmeret al. 245

Table 2. Summary of predation on radio-equipped adult mallards and wood ducks occurring on the study area, 1968-72.

Average Number of Daily Estimated Sample tracking predator predation percent Species Sex size period (days)a kills rate (m)b mortalityc Mallard F 85 43.3 9 0.0024 25.0 M 33 22.0 0 0 0 Wood duck F 62 56.3 10 0.0029 29.4 M 31 36.9 1 0.0009 10.2 a Period from the time of instrumentation until the last day the transmitter signal was detected or until the predator- killed bird was discovered. b Daily predation rate is the number of kills c predator divided by the total tracking period of all birds in the sample. Estimated percent mortality during a 120-day breeding season. Calculated using the formula: (1-(1-m)120) X 100. packages for both species averaged 23 tions were made during all hours when grams. The latter package weights were disturbance to the instrumented birds could 2.0 and 3.5 percent of the average body be minimized. If a bird had not moved weights for adult females of these respec- noticeably during the last 24 hours and the tive species bagged by Minnesota hunters signal indicated inactivity, the bird's status (Marshall and Harris 1953). was investigated. Patagial tags (Ball 1971:7) were also Hunters who reported recovering radio- placed on radio-equipped birds during 1968 equipped ducks were contacted by mail and and 1969. All ducks were fitted with U.S. given a brief history of the bird. We also Fish and Wildlife Service leg bands. Each enclosed an addressed envelope and a ques- time a bird was captured, its general phys- tionnaire requesting information on: (1) ical condition was recorded. Recaptured the location and date of recovery, (2) the birds were examined for feather wear or number of birds in company with the radio- skin abrasion around the transmitter. In marked bird, (3) feather wear or skin some cases, recaptured birds were re- abrasion due to the radio-package, (4) equipped with a new radio package. whether the hunter thought the flight of Observations on the behavior of both the bird was impaired by the radio-package, radio-equipped and unmarked ducks were and (5) the general condition of the bird recorded on data sheets by means of nu- when shot. merical codes describing 29 different types of behavior. These types consisted of six RESULTSAND DISCUSSION basic categories: (1) maintenance, (2) Predation mobility, (3) breeding, (4) social, (5) nest- We had an to evaluate ing and brood rearing, and (6) disturbance. opportunity pre- Each time a bird was observed we recorded dation because transmitters often helped us find dead birds. Predation was undetected up to three behavioral situations. During if: (1) the bird left the area before being a single observation period, several sets of killed, (2) the transmitter failed before the behavioral situations were usually recorded. kill, or (3) the predator rendered the trans- Observations of unmarked birds were made mitter inoperative or disposed of the dead hours while during daylight conducting bird in such a manner that the signal was and trapping radiotracking operations. Lo- not detectable. It was not possible to de- cations of radio-equipped ducks were ob- termine which of these events had occurred tained about three times per day. Observa- when radio contact was lost. Among birds

J. Wildl. Manage. 38 (2):1974 246 EFFECTS OF RADIOS ON WILD DUCKS * Gilmer et al.

Table 3. Comparison of direct recovery rates for normal-banded and radio-equipped adult mallards and wood ducks captured on the study area, 1968-72.

Number banded Recovery rate %a Chi-square Species Sex Normal Radio Nonnal Radio value Mallard F 52 85 11.5 28.2 4.33* M 9 33 0 24.2 1.35 Wood duck F 47 62 14.9 8.1 0.67 M 114 31 4.4 12.9 1.75

a Includes only birds shot or found dead before, during, or immediately following the first period of migratory move- ment following banding. * P < 0.05. equipped with radios, known predation ac- breeding in North Dakota, 1963-72 (D. H. counted for 7.6 percent of the mallards and Johnson and A. B. Sargeant, personal com- 12.0 percent of the wood ducks. munication). Schladweiler and Tester (1972) Female ducks were usually tracked for compared predation on radio-equipped ver- longer periods than were males because all sus banded, pen-reared game farm mallards of the mallard males and many of the wood released into natural habitats and found no duck males departed the study area by late significant difference in rates. June. A summary of predation on radio- Radio-equipped birds probably undergo equipped adult mallards and wood ducks a period of adjustment to the package dur- is shown in Table 2. Predation rate was ing which time they are preoccupied with calculated as the number of kills per the instrument and may be predisposed to duck tracking day. Mallard predation rates predation. Most of the ducks we instru- ranged from a low of 0.0014 in 1970 and mented appear to adapt to the package 1971 to a high of 0.0048 in 1972. The lowest within 1 day; however, a more subtle ad- wood duck predation rate was 0.0008 in justment probably took place over a much 1970 and the highest was 0.0136 in 1968. longer period. One-third of the observed High wood duck predation rates in 1968 predation occurred within 10 days after may have been related to the larger and marking, and 60 percent occurred within heavier radio packages used that year. the first 3 weeks. Possible adverse effects Keith (1961:44) estimated adult mortality of trauma associated with trapping and of nesting ducks near Brooks, Alberta at 8 handling could not be evaluated. Tester percent for females and 2 percent for males. (1971:393) reported that the effect of carry- Nearly two-thirds of the breeding popula- ing a radio on both birds and mammals tion consisted of dabbler species. He de- may last for as long as 1-2 weeks after rived these estimates from the number of attachment. Dumke and Pils (1973:46) carcasses found during 2 years of intensive studying mortality of radio-equipped wild work on his study area; he considered his pheasants reported that individuals adjusted estimates to be conservative. Our estimates to the radio within 1-2 days. for the overall loss to predators during the breeding season are higher but probably less Recoveries conservative than Keith's. Mortality rates Forty-one radio-equipped adult mallards comparable to ours were estimated for an and wood ducks were reported as direct unmarked population of female mallards recoveries (killed or found dead before,

J. Wildl. Manage. 38 (2):1974 EFFECTS OF RADIOSON WILD DUCKS* Gilmer et al. 247

Table 4. Comparison of direct recoveries on and off the study area for normal-banded and radio-equipped adult mal- lards and wood ducks, 1968-72.

Recovery ratea Number recovered % on area % off area Chi-square Species Sex Normal Radio Normal Radio Normal Radio value Mallard F 6 24 33.3 45.8 66.7 54.2 0.01 M 0 8 0 100.0 Wood duck F 7 5 42.9 80.0 57.1 20.0 0.48 M 5 4 20.0 0 80.0 100.0 0.01

a Includes only birds shot or found dead before, during, or immediately following the first period of migratory move- ment following banding. during, or immediately following the first and the rate at which bands are reported. period of migratory movement following We believe the latter is probably higher for banding) between 1968 and 1972. These radio-equipped birds because of the novelty birds represent 19.5 percent of the num- of the radio to hunters. ber instrumented. Two additional radio- We compared the direct recoveries on equipped mallards were reported as indirect and off the study area for normal-banded recoveries (killed or found dead during or and radio-equipped birds and found no after the second migration season following evidence that these rates were different banding). The direct recovery rate for non- (Table 4). instrumented birds during the same period Locations of direct recoveries for radio- was 8.1 percent. equipped and normal-banded ducks were We compared the direct recovery rate compared. Although samples were small, for radio-equipped and normal-banded mal- direct recovery patterns were similar. In- lards and wood ducks (Table 3). Recovery strumented mallard recoveries were re- rates for instrumented birds were higher for ported from: Minnesota, 19; Illinois, 4; all separate species-sex combinations except Arkansas, 2; and one each from Iowa, Mis- female wood ducks. A significant difference sissippi, Nebraska, Tennessee, Wisconsin, between instrumented and normal-banded Alberta, and Manitoba. Recoveries of radio- birds was detected only for female mallards. equipped wood ducks were reported from: Comparison of recovery rates is complicated Minnesota, 6; and one each from Arkansas, by the fact that the majority of the radio- Illinois, and Mississippi. Travels of ducks equipped ducks were originally captured prior to southward migration can sometimes during the spring, while most of the normal- be extensive as indicated by the direct re- banded ducks were captured during the covery of a radio-equipped male mallard summer. Radio-equipped ducks were, there- nearly 1600 km northwest of the study area. fore, subjected to natural mortality for a Duck hunters who recovered radio- longer period prior to the hunting season equipped birds were a source of additional than were the normal-banded birds. A information. Of 43 hunters contacted, 41 higher direct recovery rate for radio- returned questionnaires. Of 38 hunters who equipped than normal-banded birds does commented on the gregariousness of marked not necessarily imply higher hunting mor- birds, 50 percent indicated that the radio- tality, because the recovery rate is the equipped bird was in a flock of two or more product of the rate at which birds are shot birds. Thirty-three percent of the respond-

J. Wildl. Manage. 38 (2):1974 248 EFFECTSOF RADIOSON WILD DUCKS Gilmer et al. ?

Table 5. Comparison of the rate at which normal-banded In addition to recapture of six hens that and radio-equipped adult female mallards and wood ducks returned to the study area 1 year after were recaptured on the study area in subsequent years, we made several ob- 1969-72. being instrumented, servations of returning radio-equipped ducks Number Percent An instru- banded recaptured that we were unable to capture. Chi-square duck hen nested and raised Species Normal Radio Normal Radio value mented wood a brood in 1970. This duck was recaptured Mallard 52 85 3.8 3.5 0.14 the following year and the old transmitter Wood duck 47 62 6.4 3.3 1.09 replaced. It nested and successfully raised Total 99 147 5.1 3.4 0.10 another brood that year. Birds returning in subsequent years after being radio-equipped had undergone at least two migratory flights ing hunters felt that some feather wear or in order to return to the area where they one re- skin abrasion was present, but only had nested previously. The rate at which ported a severe condition. Eighty-six per- radio-equipped females were recaptured in cent of the 37 hunters who commented on subsequent years was compared with that the flight of radio-equipped birds indicated for normal-banded females (Table 5). The that it appeared to be unimpaired, but a differences between these rates were not few reported abnormalities. They noted significant, although the recapture rate for "difficulty in climbing" during the flare, instrumented ducks was slightly lower than and flight that appeared "weak," or "very expected. slightly impaired." One hunter wrote that the mallard he killed lagged behind its com- Behavioral Observations panion in rising from a corn field; however, Records of observations of unmarked when this bird was we trapped originally, birds on the study area were kept primarily noted that it had a badly mangled right for information on breeding chronology foot. of the Eighty-four percent respondents and brood survival; however, we hoped that the condition of the thought general that comparisons of the behavior recorded bird shot was or excellent. they good for radio-equipped versus unmarked birds would provide a reliable evaluation of pos- Recaptures sible effects of the radio packages. Several A total of 51 recaptures (45 repeats, and important biases related to data gathering 6 returns in subsequent years) involving 39 techniques made interpretation of such different radio-equipped ducks was made comparisons difficult. By knowing the loca- during five field seasons. Most of the birds tion of radio-equipped ducks, we could recaptured during 1968 had noticeable usually make visual contact without alerting feather wear and some skin abrasion caused them. As a result, behavioral categories in- by the body loop; however, improvements dicating disturbance (alert, escape, flush) that year in the package design and fit re- accounted for only 23.9 percent of the visual duced the occurrence of feather wear and observations on radio-equipped birds versus skin abrasion in subsequent years. The only 41.4 percent for unmarked birds. We con- noticeable change caused by the transmitter sequently eliminated these observations for package in birds recaptured 1 year after all birds, leaving 3608 observations of be- marking was a slightly calloused area pos- havioral situations for radio-equipped birds terior to the axilla. and 1525 for unmarked birds. Wood duck

J. Wildl. Manage. 38(2):1974 EFFECTSOF RADIOSON WILD DUCKS* Gilmeret al. 249

Table 6. Comparison of the frequency of three types of behavior observed in radio-equipped and unmarked adult mal- lards and wood ducks on the study area, 1968-72.

Observations of radio-equipped Observations of birds unmarked birds Behavior Chi-square Species type Number % of total Number % of total value Mallard Feed on water 565 17.1 207 15.7 1.15 Feed on land 78 2.4 3 0.2 23.61** Comfortmovements" 409 12.4 34 2.6 102.90** Wood duck Feed on water 55 18.4 37 17.8 0.01 Feed on land 5 1.7 0 0 2.01 Comfortmovementsa 17 5.7 4 1.9 3.48

a Includes preen, stretch, and shake. ** P < 0.01. observations represented 8 and 14 percent nificantly more than unmarked birds. This of these respective totals. The relatively situation is similar to the apparent avoid- low number of wood duck observations was ance of water and is probably due to the due to the fact that the species tends to be fact that birds with transmitters were more more secretive than mallards which made easily located on land than were unmarked observation difficult. birds. Comfort movements (preen, stretch, Greenwood and Sargeant (1973), during and shake) were performed at a greater periodic observations, noted an average of rate by instrumented than unmarked mal- 22 percent of their mallard controls oc- lards although this rate did not change sig- cupied ponds compared with only 3 percent nificantly in wood ducks. We believe that of the mallards with radio packs. They radio packages did increase the rate of com- interpreted this as an indication of pond fort movements as Greenwood and Sargeant avoidance by radio-equipped mallards. We (1973) suggested. Most radio-equipped observed mallards "on the water" in 46.5 birds preened around and tugged at the and 76.4 percent of the behavioral situations transmitter and harness a disproportionate for instrumented and unmarked birds, re- amount of time, particularly during the first spectively (P < 0.01). For wood ducks the few days after instrumentation. percentages were 47.9 and 75.0, respectively The number of times radio-equipped (P < 0.01). However, we do not feel ducks were observed feeding on water, that these differences necessarily indicated feeding on land, and engaging in comfort avoidance of water by the radio-equipped movements during the 14-day period imme- ducks, but rather the transmitters permitted diately after the attachment of the radio us to locate and observe birds that would package and during the open-ended period not normally be readily visible in habitat more than 14 days after instrumentation such as vegetated shorelines, muskrat lodges, were compared (Table 7). Although rates and sedge mats. for feeding on water increased and comfort Table 6 shows a comparison of the fre- movements decreased as mallards adjusted quency of feeding behavior and comfort to the package, preening rates probably movements between instrumented and un- never returned to normal levels in instru- marked birds. These data indicate that mented birds. radio-equipped mallards fed on land sig- We considered the possibility that radio-

J. Wildl. Manage. 38 (2):1974 250 EFFECTS OF RADIOSON WILD DUCKS* Gilmer et al.

Table 7. Comparison of the frequency of three types of behavior observed in radio-equipped adult mallards and wood ducks on the study area during the periods 0-14 days and more than 14 days after instrumentation, 1968-72.

Observations within Observations after 14 days 14 days of instrumentation of instrumentation Behavior Chi-square Species type Number % of total Number % of total value Mallard Feed on water 292 14.1 273 22.2 35.26** Feed on land 42 2.0 36 3.0 2.34 Comfort movementsa 313 15.1 96 7.8 37.14** Wood duck Feed on water 31 18.9 24 17.8 0.01 Feed on land 3 1.8 2 1.5 0.05 Comfort movementsa 12 7.3 5 3.7 1.19

a Includes preen, stretch, and shake. ** P < 0.01. equipping nesting or brooding hens might unless the bird carried a visible marker or result in an increased incidence of nest or had some distinct morphological feature. brood abandonment. The occurrence of In some cases radio-equipped birds tended nest abandonment could not be accurately to be solitary; however, most of these in- assessed because we did not always know dividuals were known or suspected to be the breeding status of hens prior to capture. molting birds or incubating hens. Gilmer However, when abandonment was known (1971:67-70) indicated that radio-equipped to occur the apparent cause could not be mallards used home ranges comparable to related to the effect of the radio package. those reported for mallards by other in- Ball (1973:53-55) reported that the aver- vestigators. age duration of the bond between radio- Effect of instrumentation on the pair equipped mallard and wood duck hens and bond was difficult to evaluate because the their broods were comparable to the find- trauma of trapping and handling could in- ings of other investigators. fluence the status of the bond. Also, in most Dumke and Pils (1973:46) felt that an cases birds were captured as individuals instrumented pheasant was adapting to the or in groups of three or more which pre- radio when it rejoined and began moving cluded defining a pair at the time of cap- about with other birds. Most ducks ture. However, a male and a female were equipped with radios were later observed occasionally captured together and were at least once as a member of a pair or other suspected to be paired. These birds were type of social group. Interactions between usually both instrumented and released to- instrumented and other ducks in social gether. Of the 13 "pairs"of mallards radio- groups gave no indications that marked equipped, 11 remained together one or more birds were rejected or behaving abnormally. days after release. One pair separated im- We observed radio-equipped birds in all mediately after release and the other pair the behavioral situations that were observed departed the study area within 1 day. Eight in unmarked birds, such as courtship, ag- of 14 wood duck "pairs" were located to- gression, copulation, brooding young, dis- gether one or more days after instrumenta- traction display, and sitting in trees (wood tion. Three pairs separated but remained ducks). Instrumented ducks were often im- in the study area and one member of each possible to identify from other individuals of the other 3 pairs left the area shortly

J. Wildl. Manage. 38(2):1974 EFFECTS OF RADIOS ON WILD DUCKS * Gilmer et al. 251 after release. Individuals that remained in birds found from a noninstrumented popu- the study area were usually "re-paired" or lation will be more conservative and pre- in some other type of social group during dation may vary greatly because of other subsequent observations. factors such as predator abundance and Ramakka (1972) reported abnormal breed- habitat. For wood duck females, a lower- ing behavior in male woodcocks (Philohela than-expected rate of direct recovery and minor) equipped with radio transmitters. rate of recapture in subsequent years indi- Instrumented mallards and wood ducks cated that radio packages may have had were observed in courtship related behavior some effect on the long-term survival of on 14 occasions. However, these birds did these birds. not appear to be hindered by the package Further evaluation of the effects of radio and the frequency of these observations did packages on various species of birds is ob- not differ from that of unmarked ducks viously needed. New package designs and (P > 0.05). attachment methods should be tested on captive birds in flight pens as well as on CONCLUSIONS wild birds under field conditions. It is also that Results of this confirm the apparent biologists using telemetry paper findings have a to in of other techniques long way go apply- investigators (Cochran 1972:39, to reduce the size Dumke and Pils that the behavior ing existing technology 1973:46) and of the transmitter and of a bird is altered weight power unconditionally upon must in mind the attachment of a radio In order supply. Investigators keep package. that a that for to evaluate the of our of the package appears satisfactory findings study one sex or one avian not be of ducks we species may breeding ecology attempted tolerated the other sex or even to determine in what the by another, ways package related Tolerance of indi- biased the behavior of instrumented mal- closely species. vidual birds and care in lards and wood ducks. Based on the find- instrumenting may also be important in determining the ability ings presented, we concluded that abnormal of a bird to carry a transmitter package. behavior induced by trapping, handling, and instrumentation did not affect seriously LITERATURECITED the data collected on movements and habi- tat use of breeding individuals and hens BALL, I. J., JR. 1971. Movements, habitat use with broods. we that and behavior of wood duck (Aix sponsa) However, emphasize broods in north-central Minnesota as deter- there are many aspects of the behavior of mined by radio tracking. M.S. Thesis. Univ. radio-equipped ducks that could not be of Minnesota. 56pp. evaluated. 1973. Ecology of duck broods in a forested region of north-central Minnesota. The most clear-cut effect of breast- Ph.D. Thesis. Univ. of Minnesota. 67pp. mounted radio packages was an increase in BOAG, D. A. 1972. Effect of radio packages on comfort movements; however, our data behavior of captive red grouse. J. Wildl. sug- Manage. 36(2):511-518. gest that instrumented birds may decrease COCHRAN, W. W. 1972. Long-distance tracking the frequency of comfort movements and of birds. Pages 39-59 in S. R. Galler, K. and R. E. Belle- increase on water as become Schmidt-Koenig,G. J. Jacobs, feeding they ville, eds. Animal orientation and navigation. adjusted to the package. Predation rates National Aeronautics and Space Administra- tion. NASASP-262. in this study may appear high but it should 606pp. DUMKE, R. T., AND C. M. PILs. 1973. Mortality be noted that rates estimated from dead of radio-tagged pheasants on the Waterloo,

J. Wildl. Manage. 38(2):1974 252 EFFECTS OF RADIOS ON WILD DUCKS * Gilmer et al.

wildlife area. Wisconsin Dept. Nat. Resour. on breeding behavior of male woodcock. J. Tech. Bull. 72. 52pp. Wildl. Manage. 36(4):1309-1312. GILMER, D. S. 1971. Home rangesand habitat SCHLADWEILER,J. L., AND I. J. BALL, JR. 1968. use of breeding mallards (Anas platyrhynchos) Telemetry bibliographyemphasizing studies of and wood ducks (Aix sponsa) in north-central wild animals under natural conditions. Bell Minnesota as determined by radio tracking. Mus. Nat. Hist. Tech. Rep. 15. Univ. of Ph.D. Thesis. Univ. of Minnesota. 142pp. plus Minnesota. 31pp. Mimeogr. [22]. I- , ANDJ. R. TESTER. 1972. Survival and GREENWOOD,R. J., ANDA. B. SARGEANT. 1973. behavior of hand-reared mallards released in Influence of radio packs on captive mallards the wild. J. Wildl. Manage. 36(4):1118- and blue-winged teal. J. Wildl. Manage. 1127. 37( 1):3-9. TESTER, J. R. 1971. Interpretationof ecological JOHNSON,R. N. 1971. Comparative survival of and behavioral data on wild animals obtained radio-markedand non-radio-markedpen-reared by telemetry with special reference to errors juvenile cock pheasants released into the wild. and uncertainties. Pages 385-408 in Proc. Minnesota Dept. Nat. Resour., St. Paul. Q. Symp. Biotelemetry S. 57, CSIR, Pretoria, Prog. Rep. 31(1):23-32. South Africa. KEITH, L. B. 1961. A study of waterfowl ecology WILL, G. B., AND E. F. PATRIC. 1972. A con- on small impoundments in southeastern Al- tribution toward a bibliography on wildlife berta. Wildl. Monogr. 6. 88pp. telemetry and radio tracking. New York Fed. MARSHALL,W. H., AND S. W. HARRIS. 1953. Aid in Fish and Wildl. Restoration Proj. W- Species composition, sex and age ratios and 123-R and Rhode Island Agric. Exp. Stn., weights of some ducks bagged by Minnesota Contrib. 1439. 56pp. hunters. Flicker 25(2):54-61. RAMAKKA, J. R. 1972. Effects of radio-tagging Accepted 21 February 1974.

J. Wildl. Manage. 38(2):1974