Advancements in the use of acoustic methods to characterize, map, and assess spawning aggregations has expanded our
understanding of the reproductive biology, life histories, and stock sizes of vulnerable species (Costa et al. 2014, Fudge and
Rose 2009). The versatility of active acoustics (echosounders) has permitted the estimation of fish abundances and
biomasses in challenging environments, such as reefs and estuaries (Boswell et al. 2007), while the efficiency of passive
acoustics (hydrophones) to monitor the sounds produced by aggregating fishes has increasingly been embraced to identify
spawning areas and periods (Rowell et al. 2015, Wilson et al. 2014). However, a logical and desirable progression to
efficiently and accurately estimate fish abundances from their sounds has been hindered by the complexity of fish calling
rates, fish choruses, and acoustic propagation (Locascio and Mann 2008, Schärer et al. 2012, Schärer et al. 2014). As such,
a comparison between independent measurements of sound and abundance has been proposed as a feasible, initial method
to determine if levels of fish sound production can be used to estimate fish abundances (Gannon 2008, Rountree et al.
2006).
In this study, we compared Gulf Corvina (Cynoscion othonopterus) sound levels with simultaneous measurements of
densities from echosounder surveys recorded at its only known spawning aggregation in the Colorado River Delta, Mexico,
to investigate empirical relationships between fish sound levels and density. The Gulf Corvina is a sciaenid endemic to the
northern half of the Gulf of California that is exclusively targeted by a lucrative fishery during spawning periods, which
occur prior to the full and new moons in the months of February through June (Erisman et al. 2012). Currently, the
sustainability of the fishery remains in question. We conducted four days of surveys in the channels of the Colorado River
Delta prior to the new moons in the months of March and April 2014. Two surveys were conducted per day, one on the
outgoing and one on the incoming tide. Two hired fishing vessels were used to complete each survey; an active acoustic
echosounder was configured on the first vessel to estimate Gulf Corvina densities and abundances present during each
survey, and the passive acoustic equipment was installed on the second vessel to record the sounds produced by Gulf
Corvina. During surveys, the two vessels progressed through the survey area together to allow measurements of sound and
density to be comparable in space and time. The echosounder-equipped vessel conducted transects across the Delta
channel, while the passive acoustic vessel recorded ambient sound along each echosounder transect prior to its completion.
Echosounder data were calibrated and analyzed to estimate fish densities and abundance via echo counting. Passive
acoustic recordings were analyzed and filtered to calculate received sound levels attributable to Gulf Corvina chorusing.
Coupled measurements of fish density and sound production levels were mapped, compared, and modeled.
We observed large aggregations comprised of more than 1.5 million fish and elevated sound production levels distributed over 25 km of the delta. Fish densities were similar on incoming and outgoing tides (Figure 1). Conversely, fish sound
production levels were significantly higher on outgoing tides (Figure 2), suggesting increases in fish calling rates during
outgoing tides when spawning is known to occur. Maps of density and received sound levels over the frequency of
choruses depicted similar spatial patterns on the outgoing tides, indicating that sound levels effectively mapped the
distribution of the spawning aggregation. The relationship between sound levels and density varied within surveys but
stabilized during the two-hour period of peak spawning (e.g. high tide to two hours after high tide), resulting in an equation
to estimate densities from received sound levels during a two hour period in which calling rates were inferred to be stable.
Our results support the inclusion of active acoustics into assessments of spawning stock abundance and indicate that sound
levels can be used to estimate fish densities when relationships are scaled to the spatial and temporal dynamics of spawning
activity. Our approach is applicable to other soniferous, aggregating fishes, providing an additional method to assess and
monitor reproductive stocks using passive acoustics. We anticipate that current and future relationships between levels of
fish sound production and density will be improved through a better understanding and inclusion of additional parameters
that account for acoustic propagation, environmental conditions, and fish behavior 

https://proceedings.gcfi.org/wp-content/uploads/2020/08/70-GCFI-PDF-23.pdf