Fairy shrimp are integral components of vernal pool ecosystems, providing key food resources for migratory birds and amphibians. However, habitat degradation and land use change severely threaten the health of both vernal pools and the survival of fairy shrimp species. Branchinecta sandiegonensis  Fugate, 1993 has been particularly affected by urban and agricultural development in its small native range within San Diego County, California, USA. It is listed as an endangered species under federal laws and is actively managed for its conservation. However, despite its legal protections, current methods of population assessment require the capture and sacrifice of individuals of B. sandiegonensis in order to identify specimens to species level through microscopy. Furthermore, accurate species-level identification is especially important given the overlap of B. sandiegonensis with a common (non-threatened) sister species, B. lindahli Packard, 1883. Thus non-invasive methods to definitively distinguish these species are desirable. Here we demonstrate that eDNA metabarcoding methods are an effective non-invasive technique for monitoring fairy shrimp. We sampled a series of vernal pools with both eDNA and traditional dip net methods, finding that eDNA metabarcoding with 16S rDNA provides species-level resolution.

eDNA metabarcoding for B. sandiegonensis had a sensitivity of 99% and a specificity of 99% while dip net sampling had a sensitivity of 98.8% and a specificity of 89.2%.

We found that the two methods were concordant within the early hydroperiod of pools. Additionally, the eDNA method detected fairy shrimp in vernal pools up to 2 months after dip nets detected any adult individuals, suggesting eDNA methods detect fairy shrimp eggs, newly hatched larvae, or decaying individuals. Together these results provide resource managers a simple, cost effective, and non-invasive method for biomonitoring endangered fairy shrimp species.

eDNA metabarcoding data of 16S invertebrate reads were collected from 11 vernal pools is 2017. Samples were sequenced twice on MiSeq PE 2x300. Raw sequences were processed using the Anacapa Toolkit with standard parameters.  Taxonomy was assigned using CRUX-generated 16S rDNA metabarcode reference database. The CRUX reference database was generated from EMBL and NCBI repositories downloaded in May 2019.

Please see Gold et al. 2020 for detailed description of the methods.

VP_meta_data has a list of all sample names and associated metadata needed to process and analyze the raw sequencing files. Raw sequence data includes two demultiplexed MiSeq PE 2x300 zipped fastq files. CRUX-generated 16S reference database has a bowtie-2 formatted reference database as well as a corresponding fasta file of all sequences and matching taxonomy text file.

Metabarcode generated sequencing data are in the two parts, attached and at https://doi.org/10.5068/D1QD5M.