Use of passive acoustic monitoring to resolve spatial and temporal patterns of spawning activity for red drum, Sciaenops ocellatus, in the Neuse River Estuary, North Carolina.

Abstract

Spawning site selection by red drum within the Neuse River Estuary (NRE) was studied by locating vocalizing aggregations using traditional hydrophone surveys (involving deploying gear from a research vessel stationed over a sampling site) and submersible, autonomous sonobuoys. Sonobuoys allowed interval recording over multiple stations during a given night, and reduced the likelihood that boat noise would interfere with courtship and spawning behavior.

The field season extended from 8 July to 30 September 2003. Sampling sites were selected at random, with replacement, from four possible categories: shallow&#8212;soft bottom (sandy, < 3 m), medium&#8212;soft bottom (transitional sand to mud, 3-5 m), deep-soft bottom (primarily mud, > 5 m), and deep-hard bottom (oyster reef or shell, > 5 m). Synoptic egg sampling at a subset of acoustic sampling sites confirmed a statistically significant relationship between probability of detecting vocalizing red drum and Sciaenid-type egg presence.

On 30 dates between July and September 2003, vocalizing males were detected 43 (of 72) times with sonobuoys (multiple observations per site) and 35 (of 88) times while spot sampling (one observation per site). Most vocalizations (85 %) were observed at sites with water depths greater than 5 m, and 97 % occurred in August and September.

Vocalization activity was most prevalent between 8 August and 17 September 2003, with the majority of vocalizations detected during a four hour period beginning at sunset.Highest vocalization activity varied between 1 and 3 hours after sunset with the average occurring 1 &#189; hours after sunset over hard substrate sites, and 2 hours after sunset over soft substrate sites. I concluded that spawning aggregations were relatively spatially stationary once vocalizations began based on a high degree of vocalization persistence, and no significant shift in time of highest vocalization activity within the study area that would imply net movement of the aggregation over a given night's sampling period.

Given the relatively fixed position of vocalizing aggregations, it was determined that depth and proximity to the river mouth were the dominant factors defining distribution of aggregations in the NRE during the 2003 field season. Drum appeared to have a strong affinity for deeper water during spawning. Vocalization activity and Sciaenid-type egg presence was more prevalent near the river mouth, suggesting that spawning activity was centered in this area. Higher received levels at the hydrophone (dB re 1 &#956;Pa 1- 1000 Hz, Bandwidth, P-P, as an indicator of vocalizing red drum presence) were found at sites containing greater than 2.5 mg/l dissolved oxygen in bottom waters. Egg presence was also positively associated with bottom waters containing greater than 3 mg/l dissolved oxygen. This evidence suggests that hypoxia may limit available spawning sites in the NRE. Egg presence was significantly related to day of year, with an increase in probability of presence occurring near the 12 August full moon, suggesting a link between spawning activity and the lunar calendar.

The NRE presents a unique set of characteristics not previously noted in estuarine systems where red drum spawning has been confirmed. These include a meso-saline environment that experiences periodic water column stratification, often with hypoxic or anoxic bottom waters. Passive acoustic monitoring techniques offer a viable and non-invasive method to locate spawning aggregations, describe diel and seasonal spawning patterns, and determine affinities to particular habitats.