Search

ABSTRACT Population growth rates of cervids are sensitive to changes in adult female survival, but fawn recruitment tends to be the population vital rate most susceptible to density-dependent population-level influences. We conducted a study of fawn survival rates and related biology in northwest Crockett County, Texas, USA, during 2004-2007 in an area where white-tailed deer ( Odocoileus virginianus) and mule deer ( O. hemionus eremicus) existed sympatrically. Our fawn study was justified by the need to explain annual variation in female-to-fawn ratios apparent during field surveys. Using vaginal implant transmitters in radio-marked adult females to locate birth sites, we captured, weighed, and radio-collared fawns on 4 privately owned ranches. We monitored collared fawns daily and investigated mortalities within 24 hours of receiving a mortality signal to determine cause of death. When possible, we collected thymus glands to assess body condition. We captured 170 fawns (64 white-tailed deer; 106 mule deer), of which 145 were of known-age from collared females, and found that relatively heavy adult females birthed relatively heavy fawns. We also noted that mass gain of fawns and mass of fawns' thymus glands were very low compared to other studies of fawns. We documented mortalities: 47 by predation, 47 by sickness-starvation, 1 by conspecific trauma, and 3 undeterminable. Bobcats ( Lynx rufus) were responsible for nearly all fawn predation; thymus gland and fawn mass data indicated that bobcats did not select weaker fawns. We used proportional hazards regression to assess factors influencing fawn survival and competing risks of mortality and made comparisons between the species when possible. We censored 6 fawns from our analyses and found that half of mortalities occurred within 15 days postpartum, and 12 of the 139 known-age fawns (9%) died within 25 m of the birth site. Fawn survival was 78% at 6 days old, which was the median estimated age-at-capture of fawns from unknown adults. Annual fawn survival rate ranged from 50% for fawns born in 2004 to 17% for fawns born in 2006, with no overall effect of species. Fawn survival was greatest in 2004 when rain during the May-August late-gestation and lactation period was greatest. Mule deer fawns succumbed more to sickness-starvation than did white-tailed deer fawns. Rates of sickness-starvation and predation increased with deviance from mean annual birth dates and with reduced rain during the late-gestation and lactation period. Early fawn survival increased with relative fawn mass, and sex became important at about 30 days postpartum, after which males were at greater risk of sickness-starvation. Lunn-McNeil competing-risks models suggested some interaction between sickness-starvation and predation. Fawn mortality, mass gain, thymus mass, and weaning data indicated that the populations of both deer species were chronically stressed near a carrying capacity that fluctuated annually with rain and that adult females invested little energy in rearing fawns relative to captive females. © 2017 The Wildlife Society. [ABSTRACT FROM AUTHOR]

Lesser prairie-chickens (Tympanuchus pallidicinctus) are traditionallymonitored by spring road-based lek surveys and counts of males attending leks. Several weaknesses exist with ground-based monitoring methods such as the bias of restricting surveys to roads, unknown probability of lek detection, and man-hours required to survey large tracts of habitat. We evaluated aerial surveys to locate lesser prairie-chicken leks in Texas and New Mexico using a Cessna 172 airplane (C172), R-22 Beta II helicopter (R-22), and R-44 Raven II helicopter (R-44) during spring 2007-2008. We determined lek activity during surveys with remote cameras placed on leks and cross-referenced time on the photo frame to time on our Global Positioning System flight log. From remote cameras we found that 305 leks were available for detection during survey flights. We determined lek detectability was 32.7% (95% CI = 20.3-47.1%) in the C172, 72.3% (64.50-79.14%) in the R-22, and 89.8% (82.0-95.0%) in the R-44. We created 16 a priori logistic regression models incorporating aircraft platform, distance to lek, survey date, lek size, and lek type to explain lek detection from aerial surveys. Our top ranked model included platform, distance, and lek type (model weight; wi = 0.288). We had four competitive models and model averaged to draw inferences. Model averaging showed that detectability was generally greatest with the R-44, followed by the R-22, and lowest with the C172, with a slight deviation from this ranking at increased distances. Within our transect width, model averaging also suggested that detectability decreased as distance from the transect to the lek increased during helicopter surveys, and detectability increased as distance from the transect to the lek increased during C172 surveys. Furthermore,man-made leks weremore likely to be detected than natural leks and large leks were more likely to be detected than medium or small leks. Aerial surveys effectively locate new leks and monitor lek density, and alleviate weaknesses associated with ground-based monitoring. We recommend using the R-44 to conduct lek surveys while flying at an altitude of 15 m at a speed of 60 km/hr on sunny mornings. [ABSTRACT FROM AUTHOR]