Using light traps to assess larval fish and octopus paralarvae diversity and ontogenetic structure around Santa Catalina Island, CA

Monitoring planktonic larvae is important for understanding changes in recruitment, range shifts, and reproductive behavior. Here, we used light traps to examine the abundance, community diversity, and ontogenetic structure of ichthyoplankton and octopus paralarvae at three sites around Santa Catalina Island, California, USA across two seasons between 2018 and 2019. We identified 15,919 fish larvae and 205 octopus paralarvae using morphology supplemented by DNA barcoding. We found no differences in larval fish diversity or abundance between years or seasons. All octopus paralarvae were captured in the summer and were identified as Octopus bimaculatus. Cryptic substrate-associated fishes dominated samples, especially those from suborder Blennioidei and family Gobiidae. We conclude that light traps can be a valuable tool and useful supplement to other zooplankton sampling methods, especially in providing a more complete assessment of nearshore plankton communities and taxa which have been poorly represented in traditional, offshore monitoring programs. They may be particularly useful for studying cryptic and substrate-associated species whose larvae are often poorly described in the existing identification literature.

Methods
Sampling was conducted from July 7-25, 2018, March 6-10, 2019, and June 6-July 21, 2019. Samples were collected at three coves around Catalina Island, USA: Big Fisherman’s Cove (33.4425, -118.5033), Catalina “Cat” Harbor (33.4345, -118.5033), and Two Harbors (33.4425, -118.4975). At each site, we sampled zooplankton using custom-built cylindrical light traps modeled after Sponaugle & Cowen (1996). Traps were made from 505 μm mesh and measured 30 x 60 cm. To attract zooplankton, we hung a high-intensity 4 LED white dive light (Tektite Mark III) inside the trap. Traps were deployed in the evenings between 17:00 and 20:00 and collected the following morning after a soak time of 12-14 hours. Traps were deployed 1-4 m below the surface. To remove any potential effect of the traps themselves, we rotated which trap was used at each site every night. Within 1 hour of retrieval, traps were washed down with fresh water, transferring zooplankton into an attached 1L cod end. Zooplankton were then immediately preserved in 90% EtOH. Two weeks after initial sample collection, alcohol was changed to ensure effective sample preservation. Sample volumes were taken using the volume displacement method. We removed and counted all ichthyoplankton and octopus paralarvae from each sample using a dissecting microscope. Fragmented fish tails and fish heads were counted as separate individuals; all paralarvae were found intact and were counted individually.

We used genetic barcoding to further confirm species identifications. We selected 87 individuals for barcoding that represented commonly identified species, frequent morphotypes that did not align with any species in our identification guides, tentative identifications from taxa with known identification challenges (e.g., Scorpaenidae), or rare/improbable species. We extracted DNA from these individuals using Qiagen DNeasy Blood & Tissue kits (Qiagen, Inc.) following standard protocols for subsampled tissue. In most cases, due to small size the entire specimen was used for extraction due to the small size of the larva. We then amplified a 658 bp region of the COI mitochondrial genome using universal primers provided by Carolina Biological Supply (VF2_t1/ FishF2_t1/ FishR2_t1/ FR1d_t1). PCR reactions had a total volume of 25 μL, composed of 12.5 μL NEB Taq 2× Master Mix, 10.5 μL primer/loading dye mix, and 2 μL of template DNA. PCR thermal cycling consisted of an initial denaturation of 2 min at 94°C followed by 35 cycles of: 30s at 94°C, 30s at 54°C, and 60s at 72°C. The final extension occurred for 10 min at 72°C. Samples were cleaned and sequenced using the forward bacterial primer M13F at the UC Berkeley DNA Sequencing Facility (https://ucberkeleydnasequencing.com/). The resulting sequence files were examined for quality (Phred scores > 20 were retained) and trimmed by 50-75 bp at both ends.

We compared our light trap data with four other surveys of larval fish in the southern California Current: A subset of data from the California Cooperative Oceanic Fisheries Investigation (CalCOFI); a long-term recruitment monitoring program through the Partnership for Interdisciplinary Studies of Coastal Oceans (PISCO); and three shorter-term studies that monitored nearshore larval fish communities using net tows in southern California and southern Baja, MX (Barnett et al. 1984, Stephens et al. 1984, Avendaño-Ibarra et al. 2004).