Relative abundance of a dominant estuarine fish and spatial extent of structured habitats in Middle Marsh, North Carolina, USA

Structured biogenic habitats in estuarine and coastal landscapes (or seascapes) augment nekton species’ production; yet, landscape setting may make restored habitats functionally redundant to co-occurring habitats for some species. Few relationships between recruitment enhancement and continuous landscape metrics have been quantified, limiting our ability to incorporate functional redundancy into restoration practice.

To address this gap, we quantified two landscape metrics, percent structure of proximal habitat and near distance to co-occurring structured habitats, for experimentally restored oyster reefs in an intertidal landscape in Middle Marsh, North Carolina, USA. We then examined the relationships between each landscape metric and recruitment enhancement of juvenile pinfish (Lagodon rhomboides), a species that uses multiple biogenic habitats and has high site fidelity as juveniles.

We estimated that reefs with < 33% proximal structured habitat within 15 m and < 59% within 50 m contributed to recruitment enhancement, as defined by greater juvenile pinfish abundances at reefs than controls and no overlap in 95% confidence intervals. 

Additionally, functional redundancy, assigned according to percent structure of proximal habitat within 15 or 50 m of restored reefs, reduced estimates of nekton recruitment enhancement in this experimental landscape by 58.3% or 33.3 %, respectively. Estimates of nekton production augmented by restoration may be inflated if they have not been adjusted to account for habitat redundancy. 

Synthesis and applications: Landscape metrics offer some predictive capacity to help restoration practitioners avoid habitat redundancy for recruitment enhancement and extend beyond considering individual habitats toward integrating landscape scale processes into predictions and restoration practice. We demonstrated a straightforward process to identify where habitats may be redundant and incorporate the landscape into restoration siting decisions: delineate co-occurring structured habitat with publicly accessible orthoimagery and model juvenile nekton relationships with proximal percent structure and near distance to other structured habitats.

Methods
Data quantifying pinfish (Lagodon rhomboides) use of restored oyster reefs and control plots in Middle Marsh, Rachel Carson Estuarine Research Reserve, North Carolina, USA were collected from 1997 - 1999 for a prior publication: Grabowski, J. H., Hughes, A. R., Kimbro, D. L., & Dolan, M. A. (2005). How habitat setting influences restored oyster reef communities. Ecology, 86(7), 1926–1935. https://doi.org/10.1890/04-0690 (hereafter Grabowski et al. 2005). Fish data retained by the current study (Davenport et al., 2024) were collected via trap sampling using minnow traps (44.5 cm long x 24.3 cm diameter with 5-mm mesh screen and ~2,5 cm openings on two opposing sides) and modified Morton fish traps (0.7 m long x 0.6 m wide x 0.25 m high, with steel rebar frames and 5-mm nylon mesh walls containing two opposing 7 cm diameter tunnel openings)  that were unbaited and set monthly from August through November 1997 and April through November 1998 and1999 during the day. Additional species were captured by this sampling effort, but only L. rhomboides was retained for use in the current study.

Spatial data outlining the locations of structured habitats in the system were manually outlines as polygons from orthoimagery of detailed intertidal and subtidal habitats in Middle Marsh from April 11, 2010, March 05, 2012, and January 30, 2016 (NC OneMap, 2019). The earliest available imagery (nearest the dates of fish sampling as possible) that clearly showed the habitats and was closest to the maximum extent of seagrass was used. Mulitple years of imagery were referenced during this process given differential visibility based on solar glare and water levels. The map scale was set to 1:300 and the spatial extent of each each structured habitat (oyster reef – C. virginica, salt marsh – Spartina alterniflora, and seagrass - Halodule wrightii and Zostera marina mixed beds) in the study site was visually delineated by tracing along habitat edges and saved as a shapefile of polygons using ArcGIS Desktop Advanced version 10.4. To check for accuracy, these delineations were then compared to a subset of the study site for which habiats were delineated from aerial imagery (for seagrass, marsh edges, sand flat, and deeper channels), or mapped in summer and fall of 2011 with a Trimble Real-time Kinematic GPS (for oyster reef habitat; Fodrie, F. J., Yeager, L. A., Grabowski, J. H., Layman, C. A., Sherwood, G. D., & Kenworthy, M. D. (2015). Measuring individuality in habitat use across complex landscapes: Approaches, constraints, and implications for assessing resource specialization. Oecologia, 178(1), 75–87. https://doi.org/10.1007/s00442-014-3212-3). Reef and control plot centroids were mapped (from XY coordinates collected by Grabowski et al., 2005), and the reefs and controls drawn as 3 m x 5 m rectangular polygons to best represent their original form at the time of construction (Grabowski et al., 2005), and aligned with poles demarcating the corners of each plot that were visible on the orthoimagery.