Marshbird response to herbicide control of cattail in northwestern Minnesota

Abstract

AbstractWetlands provide essential habitat for a wide variety of wildlife species. In the once wetland‐rich Prairie Pothole Region and adjacent areas of central North America, many wetlands have been converted to agricultural production. Many remaining wetlands experience ecological change via the invasion and spread of non‐native plant species, such as non‐native narrowleaf (Typha angustifolia) and hybrid cattail (Typha x glauca), which spread aggressively and displace native vegetation, especially in large, impounded wetlands. Management of wetlands in these landscapes often includes broad‐scale herbicide application intended to break up mats of cattail and restore areas to more wildlife‐friendly conditions. Although restoration of wildlife habitat is a common goal of such management, marshbird response to invasive cattail control is poorly understood. To evaluate the effects of cattail management on wetland wildlife, we conducted standardized call‐broadcast surveys for 5 species of marshbirds at 9 study sites that included survey locations associated with areas treated with herbicide and paired areas not treated with herbicide in wetland impoundments in northwestern Minnesota, USA, using a before‐after, control‐impact study design. We surveyed American bitterns (Botaurus lentiginosus), least bitterns (Ixobrychus exilis), pied‐billed grebes (Podilymbus podiceps), soras (Porzana carolina), and Virginia rails (Rallus limicola) during the breeding season prior to herbicide application (late summer and early autumn of 2015) and during the 3 breeding seasons after herbicide application (2016–2018). We modeled species counts using a generalized linear mixed model with year‐by‐treatment interactions as fixed effects and site as a random effect. Before herbicide application, expected mean counts did not differ between treatment and control survey locations. Three years post‐treatment, we detected significant increases in expected mean counts at treatment compared to control survey locations for soras (t193 = −3.373, P = 0.020) and Virginia rails (t193 = −3.167, P = 0.037), and point estimates for all species except least bittern were higher at treatment survey locations. Overall, our results suggest that these marshbird species responded positively to herbicide control of invasive cattail and that breeding marshbirds in these and similar wetland systems may experience positive population response over a period of at least 3 years following treatment.