Factors influencing nest survival of white-tipped doves in the lower Rio Grande Valley, Texas

The white-tipped dove (Leptotila verreauxi) is a sedentary, secretive columbid that ranges from Argentina to the Lower Rio Grande Valley (LRGV) of south Texas. Since its designation as a game species in 1984, little research has been dedicated to the species’ reproductive ecology. My objectives were to (1) identify predators of white-tipped dove eggs and nestling, (2) determine how the predator community differs among land cover types, (3) examine the impact of vegetation and landscape variation in both citrus and woodland land cover types on nest survival, (4) examine the impact of temporal variation on nest survival, (5) provide information on general nest ecology that is currently lacking, (6) investigate movements of doves and examine feasibility of tracking and recapture using a GPS/VHF transmitters pilot study. During the summers of 2015 and 2016, I conducted nest searches in citrus and woodland sites in Hidalgo County, Texas. I placed real-time, infrared emitting camera systems on a subset of nests to monitor predation. I also trapped, banded, and placed backpack transmitters on doves in Estero Llano Grande State Park (ELGSP) In the 2 years, I located 63 dove nests, 34 in citrus and 29 in woodland. I placed camera systems on 33 nests and identified 9 species of nest predator. Green jays (Cyanocorax incas) were our most common nest predator, accounting for 10 of 28 predation events. Other predators were crested caracara (Caracara cheriway), Harris’s hawk (Parabuteo unicinctus), great-tailed grackle (Quisicalus mexicanus), tawny crazy ant ( Nylanderia fulva), rat (Rattus spp.), opossum ( Didelphis virginiana), house cat (Felis catus), and Texas indigo snake (Drymarchon melanurus erebennus). Based on AIC candidate model selection, I identified nest stage as the best predictor of daily nest survival rate across both land cover types. By separating land covers for a second step to model selection, I identified different environmental variables as predictors of daily survival rate in each land cover type. In the woodland site, my null model was most important, indicating that no measured variables were important for predicting nest survival. In citrus, canopy cover was the top model. In citrus, a diverse predator community due to heavy human disturbance may have increased the importance of canopy cover and other concealment variables for nest survival. The different predator communities they encounter in the two land cover types that they nest in prioritize much different environmental conditions for nest survival