Northeastern Naturalist Vol. 25, No. 4 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 587 2018 NORTHEASTERN NATURALIST 25(4):587–598 Louisiana Waterthrush Density and Productivity in Hemlock-dominated Headwater Streams: The Influence of Stream Morphology Katie B. Barnes1, Nicholas Ernst2, Michael Allen3, Terry Master 4,*, and Rabecca Lausch5 Abstract - We assessed the density and productivity of Parkesia motacilla (Louisiana Waterthrush) at 4 Tsuga canadensis (Eastern Hemlock)-dominated headwater streams representing 2 distinct stream morphologies: ravines, defined by steep banks and fast-flowing water; and benches, characterized by braided streams meandering across a flat floodplain. The Louisiana Waterthrush is an established bioindicator of headwater-stream ecological integrity used to investigate overall habitat quality. We conducted the study from 2010 to 2013 in the Appalachian Highlands of northeastern Pennsylvania. Pair density and number of fledglings produced per km of study stream were exceptionally high at the 2 bench sites compared with the literature, and were significantly higher than in ravine sites, in part due to significantly higher nest survival on benches (69%) compared to ravines (33%). Double brooding was also significantly higher on benches. Differences in density, productivity, and nest survivorship between ravines and benches in our limited study suggest that benches may be of higher quality for this species. Our results have conservation and management implications regarding threatened Eastern Hemlock ecosystems, and the topic warrants further investigation. Introduction Determination of habitat quality for birds involves relating the habitat of a species to its survival and reproduction, often referred to as habitat-fitness potential (Beerens et al. 2015, Bock and Jones 2004, Wiens 1989). There are 2 approaches to determining habitat quality: (1) measuring habitat attributes directly and (2) using bird metrics to infer habitat quality (Johnson 2007). The latter, particularly the use of reproductive metrics, is considered more effective although often economically prohibitive (Johnson 2007). The use of measureable attributes of birds as bioindicators of habitat quality has been well documented (Caro 2010, Ormerod and Tyler 1993). Obligate riparian passerine species, especially those confined to headwater streams, are suitable bioindicators given their characteristic fidelity to the 1Birmingham Audubon Society, 118 North Royal Street, Suite 505, Mobile, AL 36602. 2Rhode Island National Wildlife Refuge, 50 Bend Road, Charlestown, RI, 02813. 3Department of Ecology, Evolution, and Natural Resources, Rutgers University, New Brunswick, NJ 08901. 4Department of Biological Sciences, East Stroudsburg University of Pennsylvania, 200 Prospect Street, East Stroudsburg, PA 18301. 5Department of Biological Sciences, Northern Arizona University, Flagstaff, AZ 86011. *Corresponding author - tmaster@esu.edu. Manuscript Editor: Jeremy Kirchman Northeastern Naturalist 588 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 Vol. 25, No. 4 immediate stream vicinity and dependence on stream resources for food and nesting habitat (Buckton and Ormerod 2008, George 2004, Master et al. 2005). These species possess the hallmark of good bioindicators: they exhibit measurable variations in behavior and demography that reflect changes in their habitat, and by extension, differences in habitat quality. Nevertheless, extensive documentation of their suitability as bioindicators has been limited to relatively few species, in particular Cinclus cinclus L. (White-throated Dipper) in Eurasia (O’Halloran et al. 1990, Ormerod and Tyler 1991, Tyler and Ormerod 1994) and Parkesia motacilla Vieillot (Louisiana Waterthrush), the only obligate stream passerine in the eastern US and a robust bioindicator of headwater stream integrity (Mattsson and Cooper 2006, Mulvihill et al. 2008, O’Connell et al. 2003). In northeastern Pennsylvania, and throughout much of the Appalachian Highlands, the Louisiana Waterthrush commonly inhabits Tsuga canadensis (L.) Carr. (Eastern Hemlock, hereafter Hemlock)-dominated headwater streams (Barnes 2014, Ernst 2012, Mattsson and Cooper 2006). These habitats have unique environmental and biological features, including dense foliage almost to ground level, deep shade, highly tannic litter, and characteristic aquatic and terrestrial animal communities (Ross et al. 2004, Snyder et al. 2002). They are currently threatened by the introduced Adelges tsugae Annand (Hemlock Woolly Adelgid) with potentially dramatic consequences for associated fauna (Ellison et al. 2005, Evans 2004, Foster et al. 2014, Toenies 2017). Preferred aquatic macroinvertebrate prey from the orders Ephemeroptera, Plecoptera, and Trichoptera (EPT) are commonly associated with these streams (Barnes 2014, Muenz et al. 2006), but Louisiana Waterthrushes consume a wide variety of both aquatic and terrestrial organisms (Mattsson et al. 2009, Mulvihill et al. 2008), thus serving as a link between both ecosystems and enhancing their value as a bioindicator (Baxter et al. 2005). They are also recognized as a forest interior species sensitive to habitat fragmentation, which further enhances their suitability as a bioindicator across spatial scales from breeding territory to the entire watershed (O’Connell et al. 2003). To better understand the linkage between reproductive success and habitat quality, we intensively studied Louisiana Waterthrush abundance and productivity at 4 streams dominated by Hemlock stands moderately impacted by adelgid infestations in eastern Pennsylvania, representing 2 distinct morphologies: ravines and benches. Ravines are characterized by high-gradient, fast-flowing streams confined between steep banks, whereas benches, underlain by harder bedrock, are characterized by braided streams that meander across a relatively wide, flat floodplain. In general, benches in this region have higher avian species richness and density than ravines (Ross et al. 2004), and their Hemlocks are in somewhat better condition (Napoli 2015). Our results will contribute to understanding whether intrinsic differences exist between benches and ravines by using the density and productivity metrics of Louisiana Waterthrush to compare habitat quality between the 2 stream morphologies. Northeastern Naturalist Vol. 25, No. 4 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 589 Methods Study sites We studied Louisiana Waterthrush pairs on 4 Hemlock-dominated headwater streams in Pike County, PA, within the Appalachian Plateaus Physiographic Province from 2010 to 2013 (Fig. 1). Hornbeck’s Creek (41º11'22.19''N, 74º53'15.78''W) and Tumbling Waters (41º9'19.98''N, 74º55'7.03'W) are ravines in the Delaware Water Gap National Recreation Area characterized by shale cliffs, waterfalls, rock/cobble substrates on bedrock and exposed root overhangs along steep stream embankments. Saw Creek (41º10' 41.99''N, 75º4'29.03''W) and Spruce Run (41º10'1.96''N, 75º7'32.68''W) are benches in nearby Delaware State Forest. Bench sites have an abundance of overturned trees with exposed root mats, moss-covered rocks, woody debris, isolated pools, and gravel/mud substrates on harder bedrock interlaced with moss-covered, exposed roots. Study-stream reaches varied in length from 1.7 km to 2.8 km (Table 1). Ravine and bench lengths differed because we made an effort to keep reaches on public/ protected land to limit access and disturbance issues. All streams are located within an extensively forested landscape dissected by paved and unpaved roads, and a few scattered houses/cabins. The 2 sites with the most fragmentation (and potential edge effects) include both a ravine and a bench site: Tumbling Waters (ravine) has a paved road at its southern end and an abandoned house along a paved road within 67 m of its upper boundary. Saw Creek (bench) is surrounded by roads, both paved Figure 1. Map of study sites in northeastern Pennsylvania. Hornbeck’s Creek and Tumbling Waters are ravines located in the Delaware Water Gap National Recreation Area. Saw Creek and Spruce Run are benches located in Delaware State Forest. All sites are located in Pike County, PA. High-resolution aerial orthoimages from 2003 to 2006 are from Pennsylvania Spatial Data Access (http://pasda.psu.edu). Northeastern Naturalist 590 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 Vol. 25, No. 4 and unpaved, and a power-line cut 290 m from the stream. A narrow dirt road with several sporadically used cabins also crosses this stream. Hornbeck’s Creek (ravine) and Spruce Run (bench) are minimally impacted by 1 perpendicular (500 m) and parallel (240–950 m) road, respectively. Direct human impact is minimal on all 4 streams and largely confined to a few hikers and fishermen using trails that parallel each stream. All sites are dominated by Hemlock stands that may be nearly pure, as is characteristic of this species, or mixed sparingly with Acer spp. (maple), Carya spp. (hickory), Betula spp. (birch), Quercus spp. (oak ), Fagus grandifolia Ehrh. (American Beech), and Tilia americana L. (American Basswood) (Napoli 2015). Understory and groundcover is generally limited by shading from the Hemlock canopy, but areas exist where fallen Hemlocks and adelgid-induced defoliation have created canopy gaps, resulting in denser understory growth, consisting largely of Rhododendron maximum L. (Great Rhododendron), Hamamelis virginiana, Gronov ex L. (Witch-hazel), Lindera benzoin Boerth. ex Schaef (Spicebush), Acer rubrum L. (Red Maple), and, especially, Betula lenta L. (Black Birch) (Napoli 2015). Banding Throughout the study, we banded virtually all male Louisiana Waterthrushes on study streams with USFWS aluminum bands and a unique color sequence of up to 3 celluloid color bands (Avinet®, Portland, ME). Males responded readily to song playback and were easily captured before or shortly after pair formation. Females were less responsive to song playback making them more difficult to capture. Whenever possible, incubating females were captured off nests and uniquely marked in similar fashion to males. This allowed us to determine site fidelity, track individual movements, delineate territories, and document nest ownership as well as instances of double brooding. Pairing success, pair density, and productivity Accumulation of GPS coordinates (Garmin 60CSX®, Garmin Ltd., Olathe, KS) of singing males allowed us to determine movements and delineate territories of all males along study streams over the length of the breeding season from 1 April through 31 July each year. We visited study streams daily, alternating visits between sites. The number of waypoints, when downloaded onto maps using ArcGIS 10® Table 1. Pair density and productivity measures from 2010 to 2013. # double # failed Reach # # broods/total nests/ Daily nest length pairs/ fledglings/ successful total Exposure survival Stream Habitat (km) km km nests monitored days rate (%)* Hornbeck’s Creek Ravine 2.7 1.9 ± 0.3 6.1 ± 2.4 0/14 6/17 154.5 96.2 ± 1.1 Tumbling Waters Ravine 2.8 1.8 ± 0.1 5.3 ± 1.1 1/14 6/19 158.5 Spruce Run Bench 1.7 3.1 ± 0.4 12.2 ± 1.2 4 /18 7/24 472.5 98.7 ± 0.4 Saw Creek Bench 1.7 3.1 ± 0.4 14.3 ± 2.4 5/13 2/16 218.5 *Corresponds to 33% (ravines) and 69% (benches) overall success over the 28-d nest cycle. Northeastern Naturalist Vol. 25, No. 4 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 591 software (ESRI, Redlands, CA) clearly defined individual territories as well as segments of streams unoccupied by either paired or unpaired males. We employed 100% minimal convex polygons to determine territory area, and thus, length, which provided a clear delineation of pairs along streams necessary for nest monitoring, and determining pair density and fledgling production, as measured per km of study-stream reach, as well as extent of double brooding and nest survival. A few males never paired, and those individuals often continued to sing throughout the season within defended territories but were not included in pair-density calculations. If males paired with different females within their territories to raise second broods, we considered them the same pair for density-estimation purposes. Nest survival We conducted nest searches and subsequent nest monitoring on each study stream. We located, recorded, and mapped nests with GPS by observing adult behaviors within each territory. We monitored active nests to check the status of eggs and nestlings every 3–4 d throughout incubation and chick rearing to limit disturbance and predation risk. We recorded clutch initiation, fledging, hatching, and nest-failure dates for all nests. If nests were found after clutch completion or hatching, then we estimated clutch initiation and hatching dates by back-calculation, using the average 13-d incubation period (beginning with last egg laid) and 10-d nestling period (Mattsson et al. 2009; Mulvihill et al. 2008, 2009). We added in a 5-d egg-laying period for a total next cycle of 28 days when calculating the daily nest survival rates. Nests with too little data were not included in Mayfield calculations but if the ultimate fate was known, they were included in the nest success analysis. We considered a nest successful if at least 1 nestling fledged at the age of 10 d, the typical number of days between hatching and fledging (Mattsson et al. 2009, Mulvihill et al. 2008), and depredated if contents were missing before the expected fledge date or if dependent fledglings were not observed with adults during subsequent visits (N. Ernst, unpubl. data). Statistical analyses For presentation of pair density and productivity results, we first computed means by site (across years) and then used these values to compute means and standard errors for all 4 sites, and for bench and ravine sites separately. To test for differences by habitat type, we used repeated measures analysis of variance (RMANOVA) on the annual data because this method accounts for non-independence of repeated measurements nested within sites. We used the lme4 package in R to conduct this analysis (Bates et al. 2014, R Core Team 2017) and included a fixed effect of habitat (bench or ravine) and a random effect of year nested within site (after Bates et al. 2014). Models were estimated with restricted maximum likelihood (REML) and fixed effects were evaluated based on Type II Wald chi-square tests. More complex models, including fixed effects of year and year x habitat interaction, were excluded iteratively using likelihood ratio tests (i.e., sequential decomposition; Bates et al. 2014). We calculated overall nest survival, as well as estimates by habitat type, based on daily survival rates (Mayfield 1961, 1975) with 95% confidence intervals Northeastern Naturalist 592 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 Vol. 25, No. 4 (Johnson 1979), and extrapolated to the full nest cycle of 28 d (Mattsson et al. 2009). We eliminated from analysis nests lacking information (found late, unknown fate) (n = 20). We performed a z-test to evaluate the differences in daily nest survival rate for the 2 habitat types. We employed the Fisher exact test to compare double-brooding rates (proportion of males that raised a second brood). We analyzed territory size and unoccupied stream-length comparisons with t-tests. We set an alpha level of 0.05 for all tests. Results Pair density Mean pair density across all 4 sites was 2.5 ± 0.4 (SE) pairs/km, with higher values at bench sites (3.1 ± 0.01; n = 2) than ravine sites (1.8 ± 0.05; n = 2; RMANOVA, χ2 = 31.8, df = 1, P < 0.001; see Table 1; Fig. 2). Territory length averaged 279 m ± 43 m on benches (n = 36 territories) and 301 m ± 59 m on ravines (n = 41). Territory area averaged 1.35 ha ± 0.98 ha on benches (n = 36) and 1.12 ha ± 0.92 ha on ravines (n = 41). We detected no statistically significant differences in territory length or area between ravines and benches (t-tests). We estimated the length of unoccupied stream reaches (no territories of either pairs or unpaired males) from our territory maps,which averaged 0.41 km ± 0.35 km on benches (n = 10) and Figure 2. Summary of major differences in reproductive metrics between Hemlock benches and ravines. Northeastern Naturalist Vol. 25, No. 4 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 593 0.35 km ± 0.17 km on ravines (n = 14); there was no significant difference (t-test). We did not measure the degree of territory overlap throughout the study, although Ernst (2012) reported that mean length of overlap was 181 m for 8 bench territorial boundaries and 112 m for 7 ravine territories. Nest survival Pooled Mayfield nest-survival estimates were as follows: 55% across all sites (95% CI = 43%, 71%; n = 76 nests; 1004 exposure days), 33% at ravine sites (CI = 18%, 62%; n = 36 nests, 313 exposure days), and 69% at bench sites (CI = 54%, 88%, n = 40 nests, 691 exposure days). The difference between ravine and bench sites was statistically significant (z-score for difference in daily survival rates = 2.17, P = 0.030). Of the 36 nests of known fate at ravine sites, 24 fledged, 11 were depredated, and 1 was abandoned. Of the 40 nests at bench sites, 31 fledged, 8 were depredated, and 1 was abandoned. Productivity and double brooding Mean productivity (fledglings produced per km of stream reach per year) across all sites was 9.5 ± 2.2 (SE; n = 4), with higher productivity at bench (13.1 ± 1.0; n = 2) vs. ravine sites (5.7 ± 0.4; n = 2; RM-ANOVA, χ2 = 28.6, df = 1, P < 0.001) (Table 1, Fig. 2). Overall, 17% of pairs double brooded, with a significantly higher percentage of bench pairs (29% [9 of 31]) raising double broods compared to ravine pairs (4% [1 of 28]) from 2010–2013 (Fisher’s exact test; χ2 = 0.038, P = 0.036) (Fig. 2). Discussion Density and productivity metrics are particularly important in evaluating habitat- quality differences (Johnson 2007). We found that pair density was significantly higher on 2 benches (3.1 ± 0.7 pairs/km) compared to 2 ravines (1.8 ± 0.4 pairs/ km) in our Pennsylvania study area from 2010–2013. Our density data for benches is the highest reported for focused population studies (Mattsson et al. 2009, Mulvihill et al. 2008, O’Connell et al. 2003; Fig. 2). Territories were stacked end to end in linear fashion on both stream types even though bench habitat is generally greater in width than ravine habitat. Potential causes of differences in pair densities, including disparities in territory length, degree of territory overlap, and the length of unoccupied portions of the stream reach, were not significantly different. Thus, higher densities on benches may be due to other factors such as increased nesting substrate and/or food-resource availability, but more research is needed to determine the causes of high density on benches. Ross et al. (2004) is the only published study that distinguished differences regarding habitat and avian characteristics between Hemlock ravines and benches. Prior to their study, Louisiana Waterthrush population metrics were largely investigated on ravine or ravine-like streams or in studies where stream types were not distinguished (Craig 1985, Mattsson and Cooper 2006, Mulvihill et al. 2008). Ross et al. (2004) reported significantly more territories of Hemlock-associated species, including Empidonax virescens Vieillot (Acadian Flycatcher), Setophaga fusca Müller (Blackburnian Warbler), Setophaga virens Gmelin (Black-throated Northeastern Naturalist 594 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 Vol. 25, No. 4 Green Warbler), and Vireo solitarius Gmelin (Blue-headed Vireo), on Hemlock benches but found Louisiana Waterthrush to have a higher affinity for ravines. Previous studies on Pennsylvania streams reported Louisiana Waterthrush pair densities of 2.3 pairs/km of stream reach in eastern Pennsylvania, 1.1 pairs/ km in central Pennsylvania, and 1.8 pairs/km in western Pennsylvania, all very similar to ravine pair densities reported here (O’Connell et al. 2003). Mulvihill et al. (2008) reported pair densities of 2.6 ± 0.2 (SE) territories/km on pH circumneutral streams (n = 4) compared to 1.0 ± 0.1 territories/km (range = 0.5–1.8) on acidified streams (n = 4) in western Pennsylvania, demonstrating that habitat quality impacts pair density. High pair-densities, such as those found on benches, indicate streams with a higher percentage of EPT, the favored insect prey of Louisiana Waterthrushes (Master et al. 2005, Mattsson and Cooper 2006). Bench sites also produced significantly more fledglings per kilometer of stream reach due to a combination of higher pair densities, higher nest-survival rates, and higher rates of double brooding (Fig. 2). The number of fledglings produced per kilometer of study-stream reach was 13.1 on benches, more than double the 5.7 on ravines. The number of young fledged per kilometer of stream reach on circumneutral streams (5.0 ± 0.4 fledglings/km) was significantly higher compared to 2.6 ± 0.4 (SE) on acidified streams in southwestern Pennsylvania because of the reduced number of territories on the latter (Mulvihill et al. 2008). Twenty-nine percent of bench pairs raised a second brood, while only 3.6% of ravine pairs double brooded. This finding compares with 4.8% (4 of 83 pairs) in Georgia (Mattsson and Cooper 2006) and 5.6% (8 of 143 pairs) in southwestern Pennsylvania (Mulvihill et al. 2009). On ravines, the lone second nest was successful producing 6 fledglings, while on benches, 7 of 9 second nests successfully produced 20 fledglings. Thus, bench pairs in our study area engaged in double brooding much more frequently than previously reported for Louisiana Waterthrush. Bench pairs are apparently more successful at overcoming documented constraints limiting the occurrence of double brooding in passerine species including length of nesting season on temperate breeding grounds (Holmes et al. 1996, Monroe et al. 2008, Mulvihill et al. 2009) and food availability (Holmes et al. 1992; Nagy and Holmes 2005a, b; Rodenhouse and Holmes 1992). Based on banding information, at least 1 member of double brooding pairs was an experienced adult more than 1 y old. Older individuals in many songbird species often settle in more productive habitats and initiate nesting earlier, both characteristics that make double brooding more likely (Mahoney et al. 2001, Ortega 2006). Time is also a critical factor in the ability to double brood. The nest-initiation date was on average 8 May for ravines and 12 May for benches from 2010 to 2012 (Ernst 2012) and 3 May for both benches and ravines from 2013 to 2014 (Barnes 2014). Median fledging date for nests included in Mayfield calculations was 9 June for ravines (23 May–27 June) and 11 June for benches (23 May–30 June). Median nest-failure date was 3 June (17 May–16 June) on ravines compared to 30 May on benches (11 May–18 June). Thus, it appears there is no obvious influence of nesting chronology responsible for the higher incidence of double brooding on benches; a more likely explanation may be higher habitat-qua lity on benches. Northeastern Naturalist Vol. 25, No. 4 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 595 Predation was the predominant cause of nest failure on both stream types. From 2010 to 2013, the nest-predation rate for nests in ravine sites was 28% compared to 19% on bench sites. The results are comparable to other studies where nest predation caused the loss of 30% of all nests across Pennsylvania (O’Connell et al. 2003). Fledglings, attending adults, and/or nest contents are susceptible to predation by many predators including Corvids, snakes, and many species of mammals (Mattsson and Cooper 2009, Robinson 1987). We documented no brood parasitism by Molothrus ater Boddaert (Brown-headed Cowbird), which is similar to other Pennsylvania studies that found low incidence (O’Connell et al. 2003), but contrasts with other regional studies (Bent 1953, Eaton 1958). Predators may find it easier to discover active nests in the eroded and exposed root overhangs used by Louisiana Waterthrushes for nest sites on ravine embankments while traveling a narrow, linear corridor bounded by high streambanks and steep inclines (Baker 1978, Ernst 2012). Benches lack the confining nature of ravines because they have braided side-channels paralleling the main channel, perhaps producing less of a funneling effect of predator movements. Louisiana Waterthrushes on benches nest almost exclusively in the root overturns of naturally fallen trees. These sites appear more randomly located across a wide floodplain than are more predictably placed root overhangs on ravines, whose locations correspond to the more regular spacing of sinuous and eroding curves characterizing these headwater streams (Mattingly 2016). Nests on benches were sometimes located relatively far from the main channel (>20 m), while on ravines virtually all nests were located immediately adjacent to or within root overhangs protruding from the stream bank (Mattingly 2016). Reduced predation pressure on benches is a contributing factor to higher nesting success and productivity on benches. Differences in edge effects on predation rates among the 4 stream reaches cannot be ruled out as a contributing cause of observed predation patterns even though streams within each type appear relatively evenly matched in this regard. The Louisiana Waterthrush is a robust bioindicator whose density, reproductive metrics, and susceptibility to nest predation reflect the quality of its headwater stream habitat (Holmes et al. 1996, Mulvihill et al. 2008, O’Connell et al. 2003). Johnson (2007) stated that bird metrics, as opposed to habitat characteristics, better reflect habitat quality. For example, Bock and Jones (2004) demonstrated that density was correlated with habitat quality for birds in general. Pidgeon et al. (2006) cautioned that individual-level metrics, such as nest success or fecundity, and population-level metrics, like fledglings produced per unit area, often classify habitats differently with respect to quality. The combined individual and population-level metrics used here support the differences found between the 4 ravine and bench streams. Our sample size was small, and thus even though study duration was relatively long, our results should be interpreted with caution as it will take additional effort to better resolve differences in habitat quality between Hemlock benches and ravines. Thus, the Louisiana Waterthrush may serve as an effective surrogate species capable of distinguishing differences in habitat quality important to a range of organisms living in Hemlock-dominated habitats, Northeastern Naturalist 596 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 Vol. 25, No. 4 especially regarding management strategies, including prioritization of treatments implemented to combat adelgid infestations. Acknowledgments The Department of Biological Sciences at East Stroudsburg University of Pennsylvania deserves thanks for its support, both financial and otherwise, during the course of this project. Funding was provided by the Faculty Development and Research Committee. J. Shreiner and C. Vorhees were instrumental in obtaining permits and permissions to conduct our research in the Delaware Water Gap National Recreation Area and Delaware State Forest, respectively. Megan Napoli provided much expert assistance in the field at a time when she was busy conducting her own research on Acadian Flycatchers on Hemlock ravines and benches. The manuscript benefited substantially from the comments and suggestions of 2 anonymous reviewers. Literature Cited Baker, R.R. 1978. The Evolutionary Ecology of Animal Migration. Hodder and Stoughton, London, UK. 1012 pp. Barnes, K.B. 2014. Using Louisiana Waterthrush (Parkesia motacilla) and aquatic-insect metrics to gauge productivity in two Eastern Hemlock habitats. M.Sc. Thesis. East Stroudsburg University, East Stroudsburg, PA. 77 pp. Bates, D., M. Machler, B. Bolder, and S. Walker. 2014. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:1–48. Baxter, C.V., K.D. Fausch, and W.C. Saunders. 2005. Tangled webs: Reciprocal flows of invertebrate prey link streams and riparian zones. Freshwater Biology 50:201–220. Beerens, J.M., P.C. Frederick, E.G. Noonburg, and D.E. Gawlick. 2015. Determining habitat quality of species that demonstrate dynamic habitat selection. Ecology and Evolution 5(23):5685–5697. Bent, A.C. 1953. Life Histories of North American Wood Warblers, Part 2. Dover Publications, New York, NY. 734 pp. Bock, C.E., and Z.F. Jones. 2004. Avian habitat evaluation: Should counting birds count? Frontiers in Ecology and the Environment 2:403–410. Buckton, S.T., and S.J. Ormerod. 2008. Niche segregation of Himalayan river birds. Journal of Field Ornithology 79:176–185. Caro, T. 2010. Conservation by proxy: Indicator, Umbrella, Flagship and other Surrogate Species. Island Press, Washington, DC. 734 pp. Craig, R.J. 1985. Comparative habitat use by Louisiana and Northern Waterthrushes. Wilson Bulletin 97:347–355. Eaton, S.W. 1958. A life-history study of the Louisiana Waterthrush. Wilson Bulletin 70:210–235. Ellison, A.M., M.S. Bank, B.D. Clinton, E.A. Colburn, K. Elliott, C.R. Ford, D.R. Foster, B.D. Kloeppel, J.D. Knoepp, G.M. Lovett, and J. Mohan. 2005. Loss of foundation species: Consequences for the structure and dynamics of forested ecosystems. Frontiers in Ecology and the Environment 3:479–486. Ernst, N.T. 2012. The Louisiana Waterthrush (Parkesia motacilla) as a bioindicator of hemlock habitat productivity: A comparison of hemlock ravines and benches. M.Sc. Thesis. East Stroudsburg University, East Stroudsburg, PA. 102 pp. Evans, R.A. 2004. Hemlock Woolly Adelgid and the disintegration of Eastern Hemlock ecosystems. Park Science 22:53–56. Northeastern Naturalist Vol. 25, No. 4 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 597 Foster, D.R., B. Baiser, A.B. Plotkin, A. D’Amato, A. Ellison, D. Orwig, W. Oswald, and J. Thompson. 2014. Hemlock: A Forest Giant on the Edge. Yale University Press, New Haven, CT. 306 pp. George, G.A. 2004. Resource partitioning and habitat use among a guild of resident and migratory riparian passerines in Costa Rica. M.SC. Thesis. East Stroudsburg University, East Stroudsburg, PA. 64 pp. Holmes, R.T., T.W. Sherry, P.P. Marra, and K.E. Petit. 1992. Multiple brooding and productivity of a neotropical migrant, the Black-throated Blue Warbler (Dendroica caerulescens), in an unfragmented temperate forest. Auk 109:321–333. Holmes, R.T., P.P. Marra, and T.W. Sherry. 1996. Habitat-specific demography of breeding Black-throated Blue Warblers (Setophaga caerulescens): Implications for population dynamics. Journal of Animal Ecology 65:183–195. Johnson, D.H. 1979. Estimating nest success: The Mayfield method and an alternative. Auk 96:651–661. Johnson, M.D. 2007. Measuring habitat quality: A review. The Condor 109:489–504. Mahoney, N.A., W.M Vander Haegen, B.L. Walker and, P.G. Krannitz. 2001. Male incubation and multiple brooding in sagebrush Brewer’s Sparrows. Wilson Bulletin 113:441–444. Master, T.L., R.S. Mulvihill, R.C. Leberman, and J. Sanchez. 2005. A preliminary study of riparian songbirds in Costa Rica with emphasis on wintering Louisiana Waterthrushes. Pp. 528–532, In C.J. Ralph (Ed.). Bird Conservation, Implementation and Integration. 3rd International Partners in Flight Conference, Volume 1. General Technical Report PSW-GTR 191, USDA Forest Service, Pacific Southwest Research Station, Albany, CA. 651 pp. Mattingly, J. 2016. Limiting factors and their role in explaining differences in Louisiana Waterthrush (Parkesia motacilla) reproductive success between hemlock ravines and benches. M.Sc. Thesis, East Stroudsburg University of Pennsylvania. East Stroudsburg, PA. 79 pp. Mattsson, B.J., and R.J. Cooper. 2006. Louisiana Waterthrushes (Seiurus motacilla) and habitat assessments as cost-effective indicators of in-stream biotic integrity. Freshwater Biology 51:1941–1958. Mattsson, B.J., T.L. Master, R.S. Mulvihill, and W.D. Robinson. 2009. Louisiana Waterthrush (Seiurus motacilla). No. 151, In A. Poole (Ed.). The Birds of North America online Cornell Lab of Ornithology, Ithaca, NY. Available online at https://birdsna.org/ Species-Account/bna/home. Accessed 5 March 2018. Mayfield, H. 1961. Nesting success calculated by exposure. Wilson Bulletin 73:255–261. Mayfield H. 1975. Suggestions for calculating nesting success. Wilson Bulletin 87:456–466. Monroe, A.P., K.K. Hallinger, R.L. Brasso, and D.A. Cristol. 2008. Occurrence and implications of double brooding in a southern population of Tree Swallows. Condor 100:382–386. Muenz, T.K., S.W. Golladay, G. Vellidis, and L.L. Smith. 2006. Stream-buffer effectiveness in an agriculturally influenced area, southwestern Georgia: Response of water quality, macroinvertebrates, and amphibians. Journal of Environmental Quality 35:1924–1938. Mulvihill, R.S., F.L. Newell, and S.C. Latta. 2008. Effects of acidification on the breeding ecology of a stream-dependent songbird, the Louisiana Waterthrush (Seiurus motacilla). Freshwater Biology 53:2158–2169. Mulvihill, R.S., S.C. Latta, and F. Newell. 2009. Temporal constraints on the incidence of double brooding in the Louisiana Waterthrush. The Condor 111:341–348. Northeastern Naturalist 598 K.B. Barnes, N. Ernst, M. Allen, T. Master, and R. Lausch 2018 Vol. 25, No. 4 Nagy, L.R., and R.T. Holmes. 2005a. Food limits annual fecundity of a migratory songbird: An experimental study. Ecology 86:675–681. Nagy, L.R., and R.T. Holmes. 2005b. To double brood or not? Individual variation in the reproductive effort in Black-throated Blue Warblers (Dendroica caerulescens). Auk 122:902–914. Napoli, M. 2015. Acadian Flycatcher response to deteriorating Eastern Hemlock conditions in two Hemlock dominated habitats in the Delaware Water Gap National Recreation Area and Delaware State Forest. M.Sc. Thesis. East Stroudsburg University, East Stroudsburg, PA. 163 pp. O’Connell, T.R., R. Brooks, R.S. Mulvihill, and T.L. Master. 2003. Using bioindicators to develop a calibrated index of regional ecological integrity for forested headwater ecosystems. Final report to US Environmental Protection Agency, STAR Grants Program, Report Number 2003-1. Penn State Cooperative Wetlands Center, Penn State University, University Park, PA. 95 pp. O’Halloran, J., S.D. Gribbin, S.J. Tyler, and S.J. Ormerod. 1990. The ecology of Dippers, Cinclus cinclus, in relation to stream acidity in upland Wales: Time-activity budgets and energy expenditure. Oecologia 85:271–280. Ormerod, S.J., and S.J. Tyler. 1991. Exploitation of prey by a river bird, the Dipper, Cinclus cinclus, along acidic and circumneutral streams in upland Wales. Freshwater Biology 25:105–116. Ormerod, S.J., and S.J. Tyler. 1993. Birds as indicators of changes in water quality: Upland rivers. Pp. 190–209, In R.W. Furness and J.J.D. Greenwood, (Eds.). Birds as monitors of environmental change. Chapman and Hall, London, UK. 356 pp. Ortega, Y.K., K.S. McKelvey, and D.L. Six. 2006. Invasion of an exotic forb impacts reproductive success and site fidelity of a migratory songbird. Oe cologia 149:340–351. Pidgeon, A.M., V.C. Radeloff, and N.E. Mathews. 2006. Contrasting measures of fitness to classify habitat quality for the Black-throated Sparrow (Amphispiza bilineata). Biological Conservation 132:199–210. R Core Team. 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available online at http://www.R-project. org/. Accessed 27 July 2017. Robinson, W.D. 1987. Louisiana Waterthrush foraging behavior and microhabitat selection in southern Illinois. Ph.D. Dissertation. Southern Illinois University, Carbondale, IL. 44 pp. Rodenhouse, N.L., and R.T. Holmes. 1992. Food limitations for breeding Black-throated Blue Warblers: Results of experimental and natural food reductions. Ecology 73:357–372. Ross, R.M., L.A. Redell, R.M. Bennett, and J.A. Young. 2004. Mesohabitat use of threatened Hemlock forests by breeding birds of the Delaware River Basin in northeastern United States. Natural Areas Journal 24:307–315. Snyder, C.D., J.A. Young, D.P. Lemarie, and D.R. Smith. 2002. Influences of Eastern Hemlock (Tsuga canadensis) forests on aquatic invertebrate assemblages in headwater streams. Canadian Journal of Fisheries and Aquatic Sciences 59:262–275. Toenies, M. 2017. Long-term avian community response to hemlock decline. M.Sc. Thesis. Penn State Universtiy, University Park, PA. 99 pp. Tyler, S.J., and S.J. Ormerod. 1994. The Dippers. T. and A.D. Poyser, Ltd., London, UK. 225 pp. Wiens, J.A. 1989. The Ecology of Bird Communities, Volume 1. Foundations and Patterns. Cambridge University Press, Cambridge, UK. 539 pp.