Delayed downstream detection and persistence of fish eDNA in a low-flow stream: a field observation

Technical Summary: Delayed Downstream Detection and Persistence of Fish eDNA in a Low-Flow Stream

Problem Statement
Environmental DNA (eDNA) monitoring is a critical tool for detecting species in aquatic ecosystems. However, in lotic (flowing water) systems, a significant challenge exists: eDNA detected at a specific sampling location may not indicate the local presence of the target organism. Instead, the signal may originate from upstream sources and be transported downstream. While previous research has established that downstream transport, retention, and resuspension influence eDNA detection, most studies have focused on medium-to-large river systems. There is a notable lack of field-based observations describing the temporal dynamics of downstream eDNA signals specifically within small, low-flow headwater streams. Under low-flow conditions, it is hypothesized that transported signals may persist over time and exhibit delayed detection patterns, complicating the interpretation of monitoring data.

Methodology
The study employed a controlled field observation in a small headwater stream (~1 m width) within the Yeongsan River system, Republic of Korea. The experimental reach was approximately 300 m long, characterized by low-flow conditions with mean flow velocities of ~0.1 m s⁻¹ and discharge values of ~0.002 m³ s⁻¹.

• Experimental Design: A live individual of Silurus asotus (a catfish) was placed in a submerged container at the upstream site to serve as a defined, continuous source of eDNA. After a designated observation period, the source organism was removed to test for signal persistence.
• Sampling Protocol: Water samples were collected at three locations (upstream, midstream, and downstream) over two consecutive days at nine time points (T0–T8). T0 represented the pre-release baseline; T1–T7 covered the period with the source present; and T8 represented the post-removal period.
• Laboratory Analysis: Samples were filtered through 0.45 µm membrane filters, and DNA was extracted using the DNeasy Blood & Tissue Kit. Species-specific quantitative PCR (qPCR) targeting the mitochondrial COI region of S. asotus was utilized. The assay used specific primers and a FAM-labeled probe to amplify a 251-bp fragment. Reactions were performed in triplicate with a QuantStudio 1 Real-Time PCR System.
• Quantification: eDNA concentrations were quantified as copies/L. To account for hydrological differences, discharge-adjusted signals (copies/s) were calculated by multiplying concentration by stream discharge.

Key Results

• Temporal Delay: Fish-derived eDNA signals were first detected at the upstream site and subsequently appeared at midstream and downstream locations with observable temporal delays.
• Signal Dynamics: The upstream site exhibited a relatively rapid increase in eDNA concentration following source introduction. In contrast, downstream sites showed delayed signal appearance and more gradual temporal variations. Peak signals at downstream locations occurred later than at the upstream site.
• Post-Removal Persistence: Following the removal of the source organism at T7, eDNA signals remained detectable at downstream locations during the post-removal sampling period (T8). Signals were also detected during the subsequent day of sampling (25–28 hours post-removal), indicating short-term downstream persistence of previously released eDNA under the observed low-flow conditions.
• Hydrological Context: Throughout the study, flow velocities ranged from 0.1 to 0.4 m s⁻¹, and discharge ranged from 0.001 to 0.017 m³ s⁻¹, confirming the low-flow regime.

Significance and Claims
The paper presents a field-based observation documenting the delayed downstream detection and short-term persistence of fish eDNA in a small, low-flow stream. The authors claim that these findings highlight the importance of considering downstream movement and temporal persistence when interpreting eDNA detections in lotic ecosystems. Specifically, the study suggests that in small stream systems under low-flow conditions, eDNA detected at a sampling location may partially reflect upstream sources rather than solely indicating the local presence of the target organism.

Limitations and Modesty of Claims
The authors explicitly state that the findings are derived from a single field observation under specific hydrological conditions and should be interpreted with caution. They do not claim these results are universally generalizable to all stream types or flow regimes. The study notes that while the observed patterns may reflect processes like downstream transport and temporary retention of eDNA particles (potentially associated with suspended or deposited material), these specific mechanisms were not directly examined in this study. Consequently, the authors call for additional replicated studies across broader hydrological conditions to evaluate the generality of these observations.