Zapus eDNA detection data

The New Mexico meadow jumping mouse (Zapus hudsonius luteus) is a riparian obligate species native to the southwestern United States. The species was listed endangered in 2014 due to habitat loss by over-grazing, wildfire, and recreation. Jumping mice move through streamside, herbaceous vegetation, and may leave behind cells suitable for environmental DNA (eDNA) analysis. Use of eDNA to detect the presence of the animal from vegetation can complement current survey approaches (live capturing, track plating, camera trapping) while reducing field labor and the risk of harm to the animal. We developed a highly sensitive, species-specific assay that used quantitative real-time PCR. Through dilution tests, we determined that only 3 copies of target DNA within a sample were necessary for positive detection at 95% confidence. In our first experiment, we assessed the specificity for our assays through rigorous laboratory tests involving closely related species ("Specificity test short assay" & "Specificity test long assay"). To investigate long-term persistence of eDNA on plant material ("Nest detections"), we tested our assay on 4 nests that were sampled 3 to 6 months after a radio telemetry study identified them as day-use sites. We then conducted a field survey to detect presence of New Mexico meadow jumping mice using sterile cotton swabs to sample plants at 6 locations along two occupied streams ("Field trial detections"). We collected 60 swabs along transects. We opportunistically swabbed plants (n = 9) following visual observation of jumping mice on plants. We detected eDNA in 3 of 4 nests. From transects, we detected 1 jumping mouse from a location where the animal was not previously observed. We also detected jumping mice from 3 of 9 observation informed (opportunistic) samples. We demonstrated that eDNA can persist on nests that the animal contacted long after abandonment. eDNA detection rate increased when informed by visual observation, and improvements to sampling strategies may further increase detection rate. Our results highlight the promise of using eDNA from terrestrial plants to detect presence of riparian specialists.