Nocturnal distance sampling all‐terrain vehicle surveys for nonbreeding rails

Abstract

Rails (Family: Rallidae) are among the most difficult birds to detect. Although methods have been developed to optimize detection during the breeding season, there is no current suitable survey method for the nonbreeding season. Low detection of rails and lack of suitable methods limit monitoring efforts and examination of important questions related to rail conservation and habitat management during the nonbreeding season. We present a new survey method along with suggestions for its effective use in moistsoil wetlands. We conducted nocturnal surveys during the autumns 2012-2015 in Missouri, USA, to detect sora (Porzana carolina) using hierarchical generalized distance sampling along transects that we traveled while riding all-terrain vehicles at night. We evaluated assumptions of our survey method by examining the response by radiomarked sora to survey vehicles and comparing survey counts between surveys on the same night. These surveys produced sora density estimates with error that can be used to address conservation and management questions such as habitat use and migratory timing. Published 2017. This article is a U.S. Government work and is in the public domain in the USA.

Virginia and Yellow Rail autumn migration ecology: synthesis using multiple data sets

“…Our study provides missing information that can be used to implement direct conservation and management actions, namely the active flooding of wetlands, in a way that maximizes the benefit to migratory waterbirds, including rails. While closely related species are often described as having similar characteristics, we show here that these two rails have different habitat needs than each other and different migratory timing than each other and Sora [18]. Understanding these differences is vital to the conservation and management of all rails as well as their wetland communities.…”

“…We surveyed wetlands that spanned from dry to 50 cm water depth, so we doubt water depth influenced our ability to detect Virginia Rails. We assessed this concern using radio-tagged Sora, and found that marked Sora did not run away from the approaching ATV [18] but future study is needed to understand the behavioral reaction of Virginia Rails to approaching ATVs at night using radio-marked birds. Because there are no formal population estimates for Virginia Rails and Sora we are unable to further infer differences between species abundances in our counts.…”

“…Opportunistic observations are important sources of information but not sufficient for informing conservation decisions, especially for cryptic species where specific methods (such as call broadcast surveys) are needed to increase odds of detecting an individual when it is present. https://doi.org/10.1515/ami-2017-0003 nocturnal spotlight flushing surveys under a distance sampling framework [18]. At each property, we surveyed moist soil wetland impoundments (a wetland surrounded by a levee, with manual water level manipulation; 4.5-300 ha in size; mean = 26.5 ha; Supplementary Table 1).…”

“…At each property, we surveyed moist soil wetland impoundments (a wetland surrounded by a levee, with manual water level manipulation; 4.5-300 ha in size; mean = 26.5 ha; Supplementary Table 1). We chose this method because methods developed to increase detection of rails during the breeding season (call broadcast surveys) are not effective during autumn migration since the call rate drops [18]. ATVs were driven along transects, as described in [18] and whenever a rail was observed, either flushing out of the vegetation or on the ground, a point was taken where the bird was first detected.…”

“…We chose this method because methods developed to increase detection of rails during the breeding season (call broadcast surveys) are not effective during autumn migration since the call rate drops [18]. ATVs were driven along transects, as described in [18] and whenever a rail was observed, either flushing out of the vegetation or on the ground, a point was taken where the bird was first detected. While this method detected thousands of Sora over the five-year project [4], we detected far fewer Virginia and Yellow Rails.…”