Environmental DNA (eDNA) analyses have become invaluable for detecting and monitoring aquatic and terrestrial species and assessing site biodiversity within aquatic environments or soil. Recent studies have extended these techniques by using eDNA to identify the presence of aboveground terrestrial arthropods directly from vegetative surfaces. However, while the dynamics of eDNA state, transport, and fate (its “ecology”) have been explored within aquatic environments and soil, they have yet to be explored within aboveground terrestrial systems. Here, we explore the ecology of terrestrial eDNA deposited by fluid‐feeding arthropods on leaf surfaces. We carried out a series of experiments to evaluate the optimal filter pore size for intracellular eDNA collection, how eDNA is affected by rain events, and its degradation rate under different solar radiation conditions. We found that the captured concentration of intracellular eDNA was not significantly affected by an increase in filter pore size, suggesting a wide range of viable pore size options exist for targeting intracellular eDNA. We also found extracellular eDNA from fluid excrement degrades more rapidly than intracellular when exposed to solar radiation, indicating the latter is a more viable target for collection. Finally, we identified that rainfall or mist will remove most terrestrial eDNA present on vegetation surfaces. We provide researchers and environmental managers key insights into successfully designing and carrying out aboveground terrestrial arthropod eDNA surveys that maximize detection probability.

Reptiles are increasingly of conservation concern due to their susceptibility to habitat loss, emerging disease, and harvest in the wildlife trade. However, reptile populations are often difficult to monitor given the frequency of crypsis in their life history. This difficulty has left uncertain the conservation status of many species and the efficacy of conservation actions unknown. Environmental DNA (eDNA) surveys consistently elevate the detection rate of species they are designed to monitor, and while their use is promising for terrestrial reptile conservation, successes in developing such surveys have been sparse. We tested the degree to which inclusion of surface and soil eDNA sampling into conventional artificial‐cover methods elevates the detection probability of a small, cryptic terrestrial lizard, Scincella lateralis . The eDNA sampling of cover object surfaces with paint rollers elevated per sample detection probabilities for this species 4–16 times compared with visual surveys alone. We readily detected S. lateralis eDNA under cover objects up to 2 weeks after the last visual detection, and at some cover objects where no S. lateralis were visually observed in prior months. With sufficient sampling intensity, eDNA testing of soil under cover objects produced comparable per sample detection probabilities as roller surface methods. Our results suggest that combining eDNA and cover object methods can considerably increase the detection power of reptile monitoring programs, allowing more accurate estimates of population size, detection of temporal and spatial changes in habitat use, and tracking success of restoration efforts. Further research into the deposition and decay rates of reptile eDNA under cover objects, as well as tailored protocols for different species and habitats, is needed to bring the technique into widespread use.

Recent methodological advances permit surveys for terrestrial insects from the direct collection of environmental DNA (eDNA) deposited on vegetation or other surfaces. However, in contrast to well‐studied aquatic applications, little is known about how detection rates for such terrestrial eDNA‐based surveys compare with conventional survey methods. Lycorma delicatula, the spotted lanternfly, is an emerging invasive insect in eastern North America, and a significant ecological and economic pest of forested and agricultural systems, especially grapes. During fall 2019, we conducted two rounds of paired eDNA and visual surveys for spotted lanternflies within 48 plots at 12 vineyards in New Jersey, USA. We compared detection probabilities within a multimethod occupancy modeling framework and used the results to extrapolate and inform survey design. The probability of detecting spotted lanternflies given presence in a plot was over two times higher for eDNA (84%) versus visual surveys (36%). In mid‐September, lanternfly eDNA was detected at five plots in three vineyards, while visual surveys revealed only a single individual in one plot. In early October, after dispersal of lanternflies into vineyards, lanternfly eDNA was detected in 12 plots within six vineyards compared with visual detections in six plots in two vineyards. Extrapolations based on detection and local‐scale occupancy rates indicate that only five and 12 plots would have been needed to positively detect lanternfly presence with 95% confidence using eDNA in contrast to 14 and 29 plots with visual surveys alone, respective to survey rounds. Log‐linear models revealed that visual counts of lanternflies were positively related to eDNA concentrations (R² = 71%). We provide some of the first quantitative evidence to support the enhanced sensitivity of terrestrial eDNA approaches compared with conventional methods. Such methods can augment efforts to combat invasive species through improved ability to delimit invasion fronts, identify satellite populations, and confirm local eradications.

Environmental DNA (eDNA) approaches to monitoring biodiversity in terrestrial environments have largely focused on sampling water bodies, potentially limiting the geographic and taxonomic scope of eDNA investigations. We assessed the performance of two strictly terrestrial eDNA sampling approaches to detect arboreal mammals, a guild with many threatened and poorly studied taxa worldwide, within two central New Jersey (USA) woodlands. We evaluated species detected with metabarcoding using two eDNA collection methods (tree bark vs. soil sampling), and compared the performance of two detection methods (qPCR vs. metabarcoding) within a single species. Our survey, which included 94 sampling events at 21 trees, detected 16 species of mammals, representing over 60% of the diversity expected in the area. More DNA was found for the 8 arboreal versus 8 non-arboreal species detected (mean: 2466 vs. 289 reads/sample). Soil samples revealed a generally similar composition, but a lower diversity, of mammal species. Detection rates for big brown bat were 3.4 × higher for qPCR over metabarcoding, illustrating the enhanced sensitivity of single-species approaches. Our results suggest that sampling eDNA from on and around trees could serve as a useful new monitoring tool for cryptic arboreal mammal communities globally.

Terrestrial arthropods are a diverse taxonomic group of significant ecological and economic importance. Our ability to understand the diversity that comprises this group is hampered by the variety of sampling techniques and high level of taxonomic expertise required to identify individual species. DNA metabarcoding approaches have potential to overcome these challenges but have been mainly limited to studies where DNA is directly extracted from trapped individuals. We posit that collection of environmental DNA (eDNA) deposited on vegetation surfaces could provide an alternative method of conducting metabarcoding‐based arthropod inventories. In this study, we illustrate the promise of characterizing arthropod biodiversity based on eDNA collected from terrestrial plant surfaces. We collected 40 paired samples using two novel eDNA sampling techniques—tree bark and foliage sampling—in a New Jersey, USA, pine barrens forest. Metabarcoding using two primer sets revealed significantly higher taxonomic richness for the 16S versus COI primer set (1077 vs. 650 molecular operational taxonomic units; MOTUs), as well as higher richness and diversity in foliage versus bark samples. Accumulation curves suggest that our samples captured about half of the available MOTU‐level diversity. Matching to reference databases revealed 28 arthropod orders, 181 families, 353 genera, and 292 species. Despite having lower MOTU‐level richness, the COI primer set revealed more taxa that were identified to species (197 vs. 115) and genus (227 vs. 173) thanks to a more complete reference database. The two primer sets and sampling substrates showed distinct community compositions that differed in important ecological traits (feeding guild, body size), demonstrating the utility of a multi‐faceted sampling and analytical approach. Our study highlights the value of exploiting eDNA left on plant surfaces via metabarcoding for contributing to rapid arthropod inventories, and thus realizing a range of ecological research and management goals.