Fish Out of Water?

Exploring the spread of environmental DNA through porous media

Eily 1

Eily Andruszkiewicz

What is environmental DNA?

  • Environmental DNA (eDNA) is any kind of cellular material released from an organism into its environment. It can be released in the form of feces, mucus, and tissue.
  • It can be important for tracking species, particularly invasive or endangered species, by indicating whether or not a certain species has been present in a certain area.
  • However, it is currently unknown how physical and biological conditions influence presence and movement of eDNA in fluids such as rivers or streams.

For Example: The invasive species of Asian carp are an increasing concern for the ecosystem of the Great Lakes. Scientists are hoping that eDNA can be used as a tool to monitor this species as well as others. By taking a water sample and analyzing it, if eDNA found from Asian carp is found, it would indicate that there are fish in that water body. However, because it is a novel technique, there are still many unknowns associated with it. If the sample has Asian carp eDNA, it is unknown how far upstream that water could have come from or on what timescale (days, weeks, etc). Also, it is unclear if sediments preserve or degrade eDNA and how far eDNA could travel and how different substrates (pebbles, sand) affect eDNA travel and retention.

Research questions to better understand eDNA

The main research questions of this project aimed to get a better basic understanding of eDNA and provide information for further studies to be conducted. They are: 

  1. How is eDNA transported and retained in flowing water systems?
  2. How do substrate and biological conditions affect retention of eDNA in a flowing system?

The questions were explored by both laboratory and field experiments.

Laboratory experiments



  • Pack glass column (60cm long, 4.8cm diameter) with one of four conditions: pea gravel, quartz sand, pea gravel with biofilm, quartz sand with biofilm.
  • Run peristaltic pump at 150 mL/min with conservative tracer until hits plateau to create conductivity curve.
  • Flush with clean water.
  • Run peristaltic pump at 150 mL/min with eDNA water collected from tank of bluegill until estimated plateau. Switch to clean water to flush and observe tailing behavior.
  • Collect 15 mL samples at various intervals before, during, after plateau to analyze with qPCR/ddPCR.

The set-up for laboratory experiments is pictured on the right. A: Water collected from fish tank, constantly stirred, B: Peristaltic pump, 150mL/min, C: Packed column, 4.8cm diameter, 60cm length, substrate either PG or sand with or without biofilm , D: 15mL samples collected for DNA extraction and PCR


  • Stochastic behavior for all treatments but all did completely flush.
  • Hypotheses for stochasticity include lysing of eDNA cells, sorption and desorption to substrate, size filtration (especially for sand).
  • Despite preventative measures, large variance in release solutions and close to limit of detection for qPCR.

qPCR Results from Pea Gravel are at right. Dashed line on top shows average concentration of release solution. Dashed line on bottom shows limit of detection for qPCR. Solid black line shows conductivity curve. Open circles show release solution concentrations. Green circles show effluent concentrations. After plateau and flushing with clean water effluent returns to zero, meaning all eDNA is flushed out of column or remains stuck inside column well enough not to be flushed out.

qPCR Results from Sand with Biofilm are at right. Again, after plateau and flushing with water effluent returns to zero concentration. Varying release solution as seen by the open circles. Sand compared to pea gravel does not have many concentrations close to the average release solution. This may indicate filtration by the sand causing larger eDNA particles to be filtered and trapped within the substrate.

Field Experiments


  • At Notre Dame’s Linked Experimental Ecosystem Facility (LEEF), there are four 50m long streams of varying substrate: pea gravel (PG), cobble, mixed PG and cobble, alternating PG and cobble every 2m.
  • Water collected from fish tanks containing blue gill and salt was dripped into each stream at a rate of 40 mL/min until plateau was reached based on conductivity.
  • 1 L samples were taken along each reach every 10m at plateau and then every 5 minutes at point E to observe tailing.



  • Only mixed stream saw statistically significant uptake of eDNA.
  • Long tailing effects; streams not return to initial condition.
  • Low concentration of eDNA due to dilution of the eDNA by the clean stream water.
  • Streams saw varying release solution concentrations.

Tailing behavior of mixed stream at right: eDNA concentration at end of reach after plateau. A,B,C,D,E each represent the mean(n=3) copies per microliter collected at plateau. Mean release solution(n=3):230 copies per microliter


One of the most helpful analytical tools is the use of modeling. For this project, the advection-dispersion equation was modeled with various solutions to determine the best fit for the data.

  • A one-dimensional, unsteady convective mass transfer advection-dispersion transport model with first order solute removal was used to estimate transport of eDNA in each experiment.
  • The solution to a semi-infinite, first boundary value problem with initial condition w=f(x) at t=0 and boundary condition w=g(t) at x=0 was used.
  • A code in Mathematica was written to solve for different hypotheses to compare expected results to experimental results depending on solutions to ADE model.

Advection Dispersion Equation (ADE):

where w: concentration, t: time, a: dispersion coefficient, b: pore water velocity, c: decay coefficient,  is source/sink term

First Boundary Value Problem Solution:

Solution to First Boundary Value Problem with c=0.5

Future Studies

These preliminary experiments provided useful information about the stochasticisity of eDNA and the diversity in size of eDNA. The preliminary results can be used to develop further experiments to answer remaining questions about the nature of eDNA and its motion in fluids.

The following are future research experiments designed to address areas of the past experiments that require a deeper understanding.

  • Experiments performed again with a higher concentration of eDNA.
  • Particle releases in summer 2014 at LEEF. Because the size of eDNA ranges, corn pollen and yeast will be released and counted to better understand the motion of eDNA particles of various size and act as “bookends” of the possible sizes for eDNA particles.
  • Samples will be analyzed using both quantitative PCR (qPCR) and digital droplet (ddPCR) to compare methods and increase resolution of low concentrations of eDNA, as ddPCR showed more accurate results at low concentrations of eDNA.


This project would not have been possible without the guidance and knowledge of these individuals and organizations: Dr. Diogo Bolster, Dr. Jennifer Tank, Christopher Jerde, Brett Olds, Dr. Rachel Novick, University of Notre Dame Linked Experimental Ecosystem Facility, Environmental Change Initiative


Darling, J.A., and Mahon, A.R., 2011, From molecules to management- Adopting DNA-based methods for monitoring biological invasions in aquatic environments. Environmental Research. 111(7).

Dejean, T., Valentini, A., Duparc, A., Pellier-Cuit, S., Pompanon, F., Taberlet, P., and Miaud, C,. 2011, Persistence of environmental DNA in freshwater ecosystems. PLoS ONE. 6(8).

Georgian, T., Newbold, SA Thomas, MT Monaghan, GW Mishall, and CE Cushing. 2003. Comparison of corn pollen and natural fine particulate matter transport in streams: can pollen be used as a seston surrogate? Journal of the North American Benthological Society. 22: 2-16.

Jerde, C.L., Mahon, A.R., Chadderton, W.L., and Lodge, D.M., 2011, “Sight-unseen” detection of rare aquatic species using environmental DNA. Conservation Letters. 4(2): 150-157.

Turner, C. Barnes, M.A., Xu, C.Y., Jones, S.E., Jerde, C.L., Lodge, D.M., 2014. Particle size distribution and optimal capture of aqueous macrobial eDNA. In press.



Biofilm: Group of microorganisms in which cells stick to each other on a surface

Conservative Tracer/Conductivity Curve: Conservative tracers are used to characterize residence time of water in a system. Common conservative tracers are salts and dyes. For salt, when putting salt water into a stream of clean water, the conductivity of the stream will rise as the water dilutes the salt water until a constant conductivity is reached (plateau). Once the salt water stops being added, the conductivity will drop again until only clean water is flowing through the system and there is no salt remaining.           

ddPCR: Digital droplet polymerase chain reaction. PCR is a biochemical technology that amplifies a targeted DNA molecule to create millions of copies of a particular sequence. Digital droplet PCR separates samples into droplets based on oil-water emulsion and then performs PCR in each droplet to quantify.

Desorption: Process in which substance is released from a surface.

Lysing: Process in which cells are ruptured or broken. Can be caused by physical external force or chemical condition.

Peristaltic Pump: Positive displacement pump for fluids. Fluid enters circular tubing which goes through a series rollers that are pinched closed under compression, forcing the fluid to be pumped through the tube.

Plateau: State reached when there is little or no change in concentration as time progresses.

qPCR: Quantitative polymerase chain reaction. PCR is a biochemical technology that amplifies a targeted DNA molecule to create millions of copies of a particular sequence. Quantitative PCR goes a step further to actually count copy numbers by detecting amplified DNA in “real time” as opposed to at the end in regular PCR.

Sorption: Physical and chemical process in which one substance attaches to another.

Stochastic Behavior: From probability theory, non-deterministic (random).

Tailing Behavior: Behavior that occurs after plateau has been reached and concentration begins to decline again.