Substantial pulses of aquatic insects emerge from tidal freshwaters along the James River Estuary, Virginia, USA

Tidal freshwaters in upper estuarine reaches provide important ecosystem services but are threatened by relative sea-level rise and pollution from increased development. Tidal freshwaters are highly productive and support estuarine and riparian food webs alike. Aquatic insects are common prey subsidies crossing into riparian habitats; however, the magnitude, timing, and composition of insect emergence in tidal systems has received little attention. Our objective was to better understand the magnitude and variability of aquatic insect emergence in tidal freshwaters. To do so, we quantified insect emergence from tidal creeks and estuarine shorelines of the James Estuary, Virginia, USA, and characterized spatial and temporal patterns in the amount of emergent biomass. We continuously monitored insect emergence from 7 April to 8 November 2019 with floating emergence traps to estimate daily emergence, then used generalized additive mixed models to analyze spatial and temporal variation in daily emergence rates. We estimated aquatic insect biomass to emerge at a mean rate (±1 SE) of 15.6 ± 2.0 g dry mass m −2 y −1, which is among the highest of previously published estimates from nontidal systems (mean ±1 SE = 12.9 ± 6.2 g dry mass m −2 y −1 ). Spatial variability in emergence was highly taxon specific. Diptera and Trichoptera had more biomass emerging from the subtidal than intertidal zone, Odonata biomass emerged more from tidal creeks than along the estuarine shoreline, and the amount of Trichoptera biomass increased, whereas Ephemeroptera decreased, with distance from the estuarine shoreline. The magnitude and composition of emergent taxa varied throughout the sampling period, with sequential peaks in biomass that altered the prey available to riparian consumers. Our results suggest that tidal freshwaters export substantial quantities of aquatic insects, which are valuable prey items for riparian consumers in these systems.

Methods
At each site, we placed 4 emergence traps along the estuarine shoreline near the creek–estuary confluence (hereafter, shoreline) and 9 traps longitudinally along the creek, reaching 1050 to 1540 m from the confluence. We positioned all traps within 3 m of the shore (creek or estuarine) at high tide. Variation in water depth resulted in some traps resting on exposed substrate at low tide (intertidal), whereas others always remained over water (subtidal). Additionally, variation in streambank slope resulted in different tidal zone sampling patterns between sites. At Deep Bottom, we sampled Bailey Creek primarily in the subtidal zone and the shoreline primarily in the intertidal zone. At the Rice Center, we sampled Kimages Creek primarily in the intertidal zone and the shoreline primarily in the subtidal zone.

We continuously monitored aquatic insect emergence from 7 April to 8 November 2019, capturing all emergence events within the sampling period. We constructed floating emergence traps following Cadmus et al. (2016) with a few modifications to increase stability and allow for continuous field placement in a tidal system. We used white no-see-um mosquito netting (30.5g/m2 ; Ripstop by the Roll, Durham, North Carolina) to capture small-bodied insects and reduce shading by the trap, which has been found to cause insect avoidance (Davies 1984). We tethered each trap to a 3-m metal conduit pole driven into the sediment to limit drift but allow for vertical movement with tidal changes. Traps covered a basal area of 0.4 m2 and included a collection bottle containing 50 to 100 mL of 70% isopropanol to preserve insects between field collections. We accessed traps by canoe at high tide every 3 to 7 d (mean ±1 SD: 5.1 ± 1.7 d) to collect samples, which is within the range of collection intervals from other studies (e.g., Whiles and Goldowitz 2001, Martin Creuzberg et al. 2017). To collect the most accurate estimate for large-bodied taxa known to avoid emergence traps (MacKenzie and Kaster 2004), we added to the sample large bodied insects (i.e., Odonata, Ephemeroptera) that were found within the trap net but not yet in the collection bottle. We stored samples in 70% isopropanol until processing, which began after the 1 st collection event and continued through August 2020.

We identified the following insects to order or suborder: mayflies (Ephemeroptera, suborders Schistonota and Pannota), stoneflies (Plecoptera), caddisflies (Trichoptera), dragonflies and damselflies (Odonata, suborders Anisoptera and Zygoptera), and aquatic flies (Diptera, suborder Nematocera) based on diagnostic morphological features (Thorp and Covich 2001). We recorded the number of individuals in each order or suborder, dried the insects for 48 h at 60°C in a drying oven, then recorded dry mass (DM) to the nearest 0.1 mg for each sample after equilibration to room temperature. For samples containing many dipterans (>200 ind.; 47% of samples), we counted a representative sample of 100 ind. and pooled the remaining insects. We used the DM of the representative sample and the remaining pooled insects to estimate the total number of dipterans in the sample. We then standardized emergent DM and density estimates from each sample and taxon m–2 d–1 based on trap area and collection interval (mg DM m−2 d−1 and ind. m−2 d−1 ).