Bird species that are restricted to tidal marshes during one or all of their life stages are under increasing pressure from sea-level rise. To date, most of the research focused on this group has been conducted during the breeding season despite the fact that more than half of the annual cycle is spent on wintering grounds and the high likelihood that the winter period is the most critical time for adult survival. We used a double-pass rope-drag technique to estimate the winter abundance of sharp-tailed sparrows (Ammospiza nelson and A. caudacutus collectively), seaside sparrows (A. maritimus) and marsh wrens (Cistothorus palustris) within tidal marshes of Virginia along 102 60×250 m transects between January and March 2014. We used the first pass to remove birds from the transect and the second pass was used to estimate detection probabilities. The technique was highly effective producing detection rates of 98% for sharp-tailed sparrows, 95% for seaside sparrows, and 91% for marsh wrens. We conducted three rounds of surveys and found that species-specific detection rates were comparable when we restricted our analyses to two survey rounds. Availability and abundance estimates deviated to a greater degree than detection rates when restricting data to that collected during only two rounds but confidence intervals overlapped for all three taxa, regardless of which two survey periods were used for the comparison. However, results were less precise when we restricted our analyses to two of three rounds with confidence intervals averaging 13%, 45%, and 14% larger for detection, availability, and abundance respectively. The double-pass rope-drag technique provides an effective, unbiased sampling technique to estimate winter songbird abundance in saltmarsh habitat provided that at least two rounds are used and increasing the number of survey rounds will result in more precise estimates.

We used double-pass rope-drag transects to survey sparrows and wrens within marsh sites. Transects were 60m wide and their length varied from 200–250 m depending on the available space and placement within the marsh patch. The 60m rope (0.635 cm diameter) was weighted in the middle by two plastic containers filled with stones to create additional vegetation disturbance. A transect was surveyed at approx. 0.8 m/s^-1 by three people where two were stationed on either end of the rope and the third walked <5 m behind the center of the rope. The third observer mapped the transect tracks to ensure that the survey was performed in a straight line. Once the entire transect length was surveyed (first pass) the observers immediately reversed direction to pass over the same area a second time (second pass) following the GPS track and observer footprints to ensure the exact same path was followed on the return pass. When birds were flushed, we recorded whether the bird landed within or outside the transect area. If birds landed within the transect area ahead of the rope, we monitored their landing location until the rope passed over them again and repeated the process until the bird flushed outside the transect. All birds that landed within the transect after flushing were eventually flushed off of the transect. Mapping allowed observers to distinguish between birds detected on the first pass from new birds detected on the second pass. Flushing birds were recorded by species group. Seaside sparrows and marsh wrens could be readily identified to species. Nelson’s sparrows and saltmarsh sparrows were combined into a single group referred to as “sharp-tailed sparrow” due to the difficulty in distinguishing these two species visually when birds quickly fly to concealing vegetation.