Factors affecting Western Atlantic red knots (Calidris canutus rufa) and their prey during spring migration on Virginia's barrier islands

Understanding factors that influence a species' distribution and abundance across their annual cycle is needed for range-wide conservation planning. Every year during spring migration, thousands of federally threatened (U.S.A.) and endangered (Canada) migratory Western Atlantic red knots (Calidris canutus rufa, 'red knot') use Virginia's barrier islands as stopover habitat to regain the fat required to continue flights to breeding grounds. Because the red knot completes one of the longest avian migrations in the world and relies on variable prey resources at its stopover grounds, the red knot exemplifies the challenges faced by long-distance migrant shorebirds. These challenges may be exacerbated by climate change, as long-distance migrants may be unable to adapt quickly to changing prey ranges and abundances, resulting in spatial and temporal mismatches between predators and prey. More specifically, as climate change causes ocean temperatures near Virginia's barrier islands to rise, organisms that live within the intertidal zone, like blue mussels (Mytilus edulis), are experiencing range shifts. Here, we 1) confirmed what prey red knots select in Virginia, 2) addressed the factors that affect red knot site selection, red knot flock size, and prey abundances across Virginia's barrier island intertidal shoreline during 2007 – 2018, and 3) predicted the origin of juvenile blue mussels, a key prey resource for red knots in Virginia.

To determine which prey are most available to red knots in Virginia, we collected sand and peat substrate core samples from Virginia's ocean intertidal zone and counted the number of prey in each sample. We compared these prey availability data to prey DNA data obtained from fecal DNA metabarcoding analyses on red knot feces (n = 100) collected on peat and sand substrates between 2017 – 2019. Red knots consumed prey from Orders Veneroida (clams), Mytiloida (mussels), Diptera (flies), and Amphipoda/Calanoida (crustaceans). While crustaceans were the most abundant prey on both sand (70.80% of total prey counted) and peat (74.88%) substrates, red knots selected crustaceans less than expected given their availability. Red knots selected clams and mussels, supporting their status as bivalve specialists in Virginia.

After determining which prey red knots consumed and selected in Virginia, we predicted the number of red knots using Virginia's barrier island stopover during their migratory stopover (May 14 – 27, 2007 – 2018) annually. We used confirmed prey, tide, distance to known roosts, and red knot winter counts from Tierra del Fuego to inform zero-inflated negative binomial mixed-effects regression models of red knot site selection and flock size in Virginia. We also used generalized linear mixed-effects regression models to determine how climatic and geomorphological factors affected prey abundances. Modeled red knot peak counts were highest in 2012 (11,644) and lowest in 2014 (2,792; x̄ = 7,055, SD = 2,841); the trend over time was variable but there was no evidence of a linear increase or decrease. Red knots selected foraging locations with more prey, though red knot flock size did not consistently relate to prey abundance. Tide, substrate, and water temperature affected prey availability. While different prey responded to these covariates in variable ways, prey generally were most abundant on peat banks at low tide.

Given the importance of blue mussels in the red knot's diet and distribution in Virginia, if the blue mussel's range continues to contract northward, red knots could be faced with additional fat replenishment challenges. We analyzed the variation in blue mussels from 2010 – 2018 by collecting core samples on peat banks in Virginia and counting the number of blue mussels in the cores. To approximate the origin of Virginia's juvenile blue mussels and determine how continued ocean temperature warming may further affect the blue mussel's range contraction, we conducted oxygen stable isotope (δ¹⁸O_c) analyses on 74 blue mussel shell umbos (the first portion of the shell precipitated) and shell edges (the most recently precipitated shell) to compare and predict where different portions of the shell were formed. We compared blue mussel shell compositions to δ¹⁸O_c calculated in equilibrium with regional ocean water using recorded δ¹⁸O_w data and sea surface temperature data from ocean buoys between New Hampshire and Virginia. Blue mussel abundance/core sample declined over the duration of our study (Spearman's rank correlation coefficient: ρ(rho) = -0.31, p < 0.001), with the highest abundance in 2010 (x̄ blue mussels/core sample = 537.88, SE = 85.85) and lowest in 2016 (x̄ = 34.08 blue mussels/core sample, SE = 6.96). Blue mussel umbos (x̄ δ¹⁸O_c = -0.23‰, SE = 0.12) contained more positive δ18Oc than shell edges (x̄ δ¹⁸O_c = -0.53‰, SE = 0.20), suggesting that Virginia's blue mussels originated from ocean populations in more saline and/or colder water than that within Virginia's intertidal zone. Blue mussel umbo δ¹⁸O_c were not different than δ¹⁸O_c calculated in equilibrium with regional ocean water off the Virginia and Delaware coasts, suggesting that Virginia's blue mussels originated in ocean waters between Delaware and Virginia; however, they may have originated in waters as far north as New York in some years, potentially decreasing the risk of blue mussels being completely extirpated from Virginia in the near future.

While red knots currently use spring migratory stopovers across the United States' Atlantic Coast, from Florida to New Jersey, the largest spring concentrations of knots are confined to the Delaware Bay and Virginia's barrier islands. Because these stopover grounds support large proportions of the red knot's migratory population, any changes in the factors that affect red knots at these stopover sites could have lasting implications for red knots. The blue mussel's range contraction and decline over time in Virginia, for example, is concerning from a conservation perspective. Red knots require easily accessible and abundant prey resources to efficiently replenish fat-stores needed for continued migration and breeding. Additionally, because red knots breed within a narrow period, any delays on stopover grounds could prevent red knots from breeding, even if they survive migration.

Our research demonstrates that red knots use prey abundance as a determinant when selecting foraging locations, and that peat banks, while only sporadically available across the barrier islands at mid- to low-tides, contain higher prey abundances than sand. Thus, to continue maximizing the availability of prey in Virginia, measures should continue to be taken to allow natural island migration processes that encourage the presence of both sand and peat substrates. Beach nourishment and stabilization projects are often used on coastal beaches to prevent shoreline erosion; however, such actions prevent the formation of peat banks by blocking island migration processes. A reduction in peat banks could decrease the abundance of prey available to red knots, making weight gain during the critical stopover period more challenging for red knots. Additionally, beach nourishment through sand replenishment buries invertebrate prey, potentially causing mass prey mortality and reducing shorebirds' ability to access deeply buried prey. To prevent the loss of important peat banks on these islands, and to prevent disrupting predator-prey interactions, managers should continue their ongoing focus on allowing natural processes to occur on Virginia's barrier islands.