Optimizing environmental DNA detection methods while analyzing the presence of river otters in the Northeast


Analyzing populations of species has come a long way from simple field research, with recent detection methods being based on DNA detection assays. Collection and analysis of environmental DNA (eDNA), the DNA contained in feces, urine, scales, skin, hair, or other excretes, can now be analyzed to monitor biodiversity and map population distribution with finite precision. My research focuses on amplifying target DNA, specifically of North American river otters, American beavers, muskrats, and northern raccoons, in river water samples using the recently-developed GoFish nested PCR method: I am the first to utilize this methodology in the detection of the aforementioned target species. Using designed 12S broad-range primers and MiFish species-specific primers, river water samples were tested for target DNA presence. I produced multiple novel significant findings. First, the broad-range primers were able to successfully detect mammal DNA, validating the GoFish PCR method. Next, the presence of river otter DNA, as well as that of other target mammals, was detected in the East Branch Croton River, Chipuxet River, and Byram River, supporting the design of viable primers. Overall, my study is the first of its kind to use eDNA as a method of environmental mapping of non-aquatic and semi-aquatic mammals in the Northeast, and pioneers the use of GoFish Nested PCR with non-marine mammal DNA. By identifying the population density of target mammals in various rivers in the Northeast, eDNA research is shown to aid conservation efforts as well as hoist detection methods and field research into the imminent future.

Question / Proposal

Can we improve how we detect and track living things, making it easier and more accessible with environmental DNA research?

Primarily, my study will focus on how the development of convenient and effective eDNA detection methods result in accurate detection of multiple species, accounting for obstacles such as the sparseness of target DNA being excreted. We will use these methods to investigate the North American river otter in the Northeast. My research hopes to address both the lack of data on river otter populations using traditional methodology as well as the insufficient eDNA biodiversity monitoring research with semi-aquatic and terrestrial species. The goal is threefold: firstly, we will optimize the GoFish nested PCR technique and demonstrate its validity in non-marine mammal testing; secondly, we will use our protocol to find the population density of river otters in various rivers in New York, Connecticut, and Rhode Island; and thirdly, we will test eDNA analysis methods on other semi-aquatic and non-aquatic mammals, such as beaver, muskrat, and raccoon. Ultimately, our goal is to demonstrate the use of a cutting-edge technique to give us more precise identification to have a clear understanding of the wildlife.​​​​​​​


Research using eDNA began in 2008, with the goal of detecting the presence or absence of different target aquatic species, first using fish species and amphibians in freshwater wetlands and rivers (Ficetola, Miaud, Pompanon, & Taberlet, 2008). Aquatic organisms were used as target species because of their ability to shed more eDNA than other species, so a higher chance of positive detection arose with the use of amphibious species (Ficetola, Miaud, Pompanon, & Taberlet, 2008). Ficetola et al. (2008) tested environmental samples using a primer for a frog species, Rana catesbeiana, to validate the first designed methods using eDNA in the field of population detection (Ficetola, Miaud, Pompanon, & Taberlet, 2008). Their goal was to determine how to make eDNA collection more cost-effective, time- efficient, and precise in species detection.

Thereafter, eDNA studies began to expand with the use of different species as well as varied PCR procedures such as quantitative PCR and nested PCR. In 2016, Adrian-Kalchhauser & Burkhardt-Holm used an eDNA assay to map the distribution of a group of invasive species, Ponto-Caspain gobies (Adrian-Kalchhauser & Burkhardt-Holm, 2016). This research provided necessary data for the eventual control of these invasive organisms, mapping the overpopulation of the species. Furthering the methods used in Adrian-Kalchhauser & Burkhardt-Holm’s study and past eDNA research, Weltz et al. (2017) mapped the distribution of the Maugean skate, an endangered fish species in Tasmania, contributing towards the reintroduction and further protection of the species. With the goal of method support, development, and advancement, eDNA research has evolved and expanded in its applications as a biodiversity monitoring tool.                                               

Altogether, these studies support the major aim of current eDNA research: to modify and strengthen methods of DNA detection from samples. Studies using eDNA analysis differ by altering species type, environment, location, population size, primer design, or techniques (Goldberg et al., 2016). Each study contains unique research. However, the majority of eDNA studies use a general six step procedure: sample collection, filtration, primer design, extraction/isolation, amplification, and analysis.

The GoFish nested PCR methodology, developed by Stoeckle et al. (2018) uses broad-range and species-specific amplifications in multiple rounds to create a more precise conglomerate of data. At this point, it has only been applied in a single study, using marine organisms. Being faster and more cost and time efficient than the other considered methods, GoFish amplification is used in this study, in a novel way with the use of non-marine mammals. The method is able to streamline both broad-range and species-specific primer design in an expeditious manner, possibly holding the potential to become a primary method in eDNA research (Stoeckle, Mishu, & Charlop-Powers, 2018).

Most eDNA studies are conducted using eDNA collected from bodies of water, such as rivers, because it is dispersed more, and easier to filter, test and analyze than collections from terrestrial environments. theless, non-aquatic and semi-aquatic populations have overall been understudied utilizing eDNA research, but hold as much importance in contributing to the understanding of the overall biodiversity of water bodies.

Method / Testing and Redesign

Part I. Primer Design

The list of species for sequencing was chosen using common mammal species near the areas of collection. A list of all mammals in the Northeast of the US, including species from various families, was compiled. DNA sequences for each of the species were collected using GenBank, and were analyzed for similarities using the NCBI BLAST database (Table 1). A phylogenetic tree was generated to show relation and similarities among species (Figure 1).

To design species-specific primers, the alignments generated from NCBI BLAST for river otter, muskrat, beaver, and raccoon were used to create shortened left-hand forward candidate primers and right-hand reverse candidate primers from the M13 region of each sequence that were unique to the target species, using the MEGA7 application (Table 2). Beaver, muskrat, and raccoon species-specific primers were used to test for false-positive detection from similar species in the positive control samples (river otter=RIOT, beaver=BEAV, muskrat=MUSK, raccoon=RACC).

Part II. Sample Collection

The samples were labeled in order of collection (Table 3). The rivers used in this study were chosen due to the rare sightings along the river, and the unknown river otter population location and size in the area, information provided by state wildlife biologists or through other personal communication. Three rivers were collected from, along with collection from a zoo river otter enclosure for positive control samples.

Part III. Filtration

        Two filtration protocols were used for different rivers – hand-pumped portable filtration (Protocol #1) and vacuumed in-lab filtration, using the same process but different materials.

Part IV. Extraction and Isolation

        Methods of extraction and isolation of samples proceeded for RO1–RO10 once, while samples RO11–RO18 were duplicated and tested twice. DNA was extracted from the filters using a MoBio Powersoil Kit, containing chemicals C1–C6 and PowerBead tubes (MoBio, 2014). The MoBio protocol was followed for extraction and isolation (MoBio Protocol)

Part V. PCR Amplification

        Two PCR amplifications in total were completed for each of the sample sets – one: necessary to detect DNA using the broad-range primer, and two: nested PCR, making the data 100-fold more sensitive and specific. After mixing each sample with a chemical master mix (developed using H2O and a target species primer for each of the four species), they were then placed in an Eppendorf PCR Cycler.

More on amplification protocols

For the second round of amplification, the same procedure was used, but with multiple different master mixes.

Part VI. Data Analysis

Once cycled, the samples were ready to be placed in a 2.5% agarose gel for analysis (1 g of agarose in 40 mL of 1X TBE) and were run using a PowerPak Basic at 60 V. Once reads were documented and confirmed, the remainder of positive samples were sent to an outside facility to analyze Sanger sequences of the banded samples, using M13 primers. The Sanger sequences then specified what particular species has the greatest concentration of DNA in the sample for the broad-range primer analysis, along with being used as a check for correct procedure and analysis.


Broad-range Primer Findings

I designed the 12S M13 LiS primer using MEGA7, and the cytochrome b primer was designed and used in a previous study (Padgett-Stewart et al., 2015).

Through Sanger sequencing:

  • Bands for RO1–RO6 did not come out with specific results to sequence any certain DNA of species as the majority of DNA detected, excluding RO6 with the cytochrome b primer and RO1 with the 12S primer.
  • Human DNA was detected as the majority of the DNAs in sample RO1 with 12S primer, showing the likelihood of human waste or human presence in and around the sample site.
  • Cow DNA was detected as the majority in samples RO9 and RO10 with the 12S primer, collected from the river otter bathing pool at the zoo, most likely showing that the otter is being fed cow meat or the water is contaminated with cow DNA, likely from food in other nearby zoo enclosures.
  • Stool samples did not detect DNA with the 12S primer, most likely resulting from the very low DNA yields detected from QuBit readings (Table 4), making the designed primer and the GoFish methodology likely just usable for water samples and not for stool samples, as well as possibly not being sensitive enough to detect small levels of DNA in samples after only one round of amplification.

  • River otter DNA was detected as the greatest amount of the DNA in samples RO7–RO10, which served as positive controls, banded when using the broad-range primers and only the RIOT species-specific primer.
  • For the 12S broad-range primer, samples RO7 and RO8 amplified river otter DNA, meaning there must have been much more river otter DNA in the water compared to any other vertebrate.

The finding of cow DNA in samples RO9 + RO10 does not skew results, but rather provides interesting insight towards what must be considered when creating a 12S primer, as well as about the feeding habits of river otters.

Species-specific Primer Findings

The presence of river otter DNA and beaver DNA was found in the Chipuxet River and the East Branch Croton River, but not in the Byram River. The presence of muskrat DNA was found in all three rivers, and raccoon DNA was only tested for in the Byram River, used as a negative control, and was present. Positive and negative correlation for different tested samples is shown on the gels below.

The presence of DNA in some but not all samples in one river can suggest that target species only inhabited a downstream location (as the case with the RIOT primer tested samples for the East Branch Croton River), the flow or other qualities of the river did not allow for the travel of eDNA to certain sites, small eDNA quantities degraded as they traveled downstream (as the case with the RIOT primer tested samples for the Chipuxet River), and/or the designed primer was not sensitive enough to detect a minimal amount of DNA in water more than once.



This study refined methods with the overall goal of contributing to the ecological field of population distribution mapping. The major problem addressed in this research is the lack of development and testing of population mapping methods, specifically using eDNA. The successful optimization of GoFish Nested PCR enabled the positive identification of non-marine mammals from eDNA water samples.

This validation of the GoFish Nested PCR methodology showed that it is applicable to detect aquatic, semi-aquatic, and non-aquatic species. Following the validation of GoFish Nested PCR to a wider range of species, my research provides evidence allowing future researchers to apply this novel method to map a greater diversity of populations.

Project Summary

  • The presence and absence of multiple elusive species was found in various rivers, most importantly showing the presence of river otters in the Chipuxet River and East Branch Croton River and suggesting their absence in the Byram River, contributing to overall population density measurements.
  • The lack of river otter eDNA in the Byram River was unexpected.
    • These results question the previous assumption of a river otter population there.
  • In order to confirm the absence of river otters, more water samples should be collected from multiple sites along the river.

Statewide Conservation Support

  • The population mapping findings will be especially important in assisting conservation efforts made by the New York State Department of Environmental Conservation (NYS DEC) and the United States Department of Agriculture (USDA).
  • Management strategies for river otter populations are specious because they lack dependable methodology for creating a baseline status of otter populations and monitoring the changes in the population (Frair, J., 2016).
  • My data can aid state and national organizations in targeted reintroduction through more in-depth and accurate tracking techniques.
  • Mapping populations of keystone species such as river otters is important as it can potentially aid in the conservation of biodiversity and even ecosystems as a whole.

Future Work

  • First, this research can be continued by testing my collected river water samples for additional non-marine mammal species. 
    • A database of samples can even allow for retesting of samples and mapping population changes over time.
  • Secondly, the designed primers, both species-specific and broad-range, can be used to expand the knowledge of population distribution of river otters and other mammal species beyond the Northeast by collecting many more water samples.
  • Finally, the primers designed in this study, along with the GoFish nested PCR methodology, can act as a resource for other researchers investigating the populations of North American river otters, beavers, muskrat and raccoons.
  • With a wide array of data from field studies, machine learning can even predict the influence of environmental factors such as pollution, on changes in populations.

Ultimately, this study, while focusing on specific species and rivers, is able to contribute to the overall conservation and biodiversity of the environment; not only do the findings provide valuable data on the populations of keystone species, but this study is also a key step towards mapping the distribution of animal populations all over the world.

About me

I am a high school senior at Byram Hills High School in Armonk, NY. I am a student in the three-year Authentic Science Research program, where I conducted a full research project on a scientific topic entirely of my choosing.

I have always loved experimenting and hypothesizing in science, and I have especially had interest in ecology and conservation. When I was 11, I organized two pollution cleanup events, one in my hometown and one on a nearby beach; I also participated in National Cleanup Day from age 11-13. I have even stayed in contact with the NRDC and other environmental protection organizations to try to help the environment in every way possible and initiate changes.

As a lover of environmental-based science, I was especially intrigued when I happened upon an article about environmental DNA, and I was immediately hooked by eDNA’s ingenuity and accuracy as a cutting-edge biodiversity monitoring tool.

My inspiration for becoming a researcher comes mostly from the importance and significance I see in all scientific work. Most of my family members have careers in education or business, so pursuing science is more unique and independent. I also love the cumulative nature of scientific research, where any research can be expanded and tested hundreds and thousands of times over, altering small variables to make a huge impact. Winning Google Science Fair will hopefully make the field of eDNA research expand, leading to the development of methodology and broadening the already tight-knit community of eDNA scientists.

Health & Safety


The safety protocol for Rockefeller University's Program for the Human Environment lab was followed throughout all lab research. No harmful materials were used in the experiment. The Minor Volunteer Form was filled out before any lab experimentation as well. 

To reduce the likelihood of cross-contamination of samples or chemicals, or mixture of unwanted chemicals, all reusable items (e.g. well trays, test tube trays, Pipetman, lab benches, etc.) were thoroughly cleaned with distilled water and 10% bleach solution after each use. For collection of samples, 1 L soda bottles were rinsed thoroughly with distilled water at least 20 times before using the bottles. Each water sample was collected at least 3 ft from the edges of the river, and the bottle was submerged at approximately 2 ft underwater to collect from the areas where eDNA transfer was most likely.


I worked alongside a professional lab researcher at Rockefeller University, Dr. Mark Stoeckle (stoeckm@mail.rockefeller.edu), for the entirety of my work in the lab. He also helped me understand the field and the purpose of my work through teaching me different processes along the way.

Bibliography, references, and acknowledgements

Adrian-Kalchhauser, I., & Burkhardt-Holm, P. (2016). An eDNA Assay to Monitor a Globally Invasive Fish Species from Flowing Freshwater. PLoS ONE, 11(1), e0147558. http://doi.org/10.1371/journal.pone.0147558

Carraro, L., Hartikainen, H., Jokela, J., Bertuzzo, E., & Rinaldo, A. (2018). Estimating species distribution and abundance in river networks using environmental DNA. Retrieved from http://www.pnas.org/content/early/2018/10/23/1813843115

Danzer, S. (2011). Byram River Watershed Management Plan. https://www.ct.gov/deep/lib/deep/water/watershed_management/wm_plans/byram_wbp2012att.pdf

DEC.ny (2018). River Otter. Retrieved from https://www.dec.ny.gov/animals/9355.html

Ficetola, G. F., Miaud, C., Pompanon, F., & Taberlet, P. (2008). Species detection using environmental DNA from water samples. Biology Letters, 4(4), 423-425. doi:10.1098/rsbl.2008.0118

Ficetola, G. F., Taberlet, P., & Coissac, E. (2016). How to limit false positives in environmental DNA and metabarcoding? Molecular Ecology Resources, 16(3), 604-607. doi:10.1111/1755-0998.12508

Frair, J. (2016). Status of the Reintroduced River Otter Population in New York State. Retrieved from https://portal.nifa.usda.gov/web/crisprojectpages/1010887-status-of-the-reintroduced-river-otter- population-in-new-york-state.html

Goldberg, C. S. et al. (2016). Critical considerations for the application of environmental DNA methods to detect aquatic species. Methods in Ecology and Evolution, 7(11), 1299-1307. https://doi.org/10.1111/2041-210X.12595

Google Maps. (2018). Retrieved September 17, 2018, from https://www.google.com/maps/place/Sodom Rd, Brewster, NY 10509/@41.3789779,73.6404136,12z/data=!4m5!3m4!1s0x89dd530 8d8b393cb:0xdd2f031d9aca204c!8m2!3d41.4016161!4d-73.5919619

Laramie, M.B., Pilliod, D.S., Goldberg, C.S., & Strickler, K.M. (2015). Environmental DNA sampling protocol—Filtering water to capture DNA from aquatic organisms: U.S. Geological Survey Techniques and Methods, book 2, chap. A13, 15 p., http://dx.doi.org/10.3133/tm2A13

Life Technologies (2015). QuBit dsDNA HS Assay Kits. Catalog No. Q32851 & Catalog No. Q32854. https://assets.thermofisher.com/TFSAssets/LSG/manuals/ Qubit_dsDNA_HS_Assay_UG.pdf

MoBio Laboratories, Inc. (2014). PowerSoil DNA Isolation Kit. Catalog No. 12888-50 & Catalog No. 12888-100. https://mobio.com/media/wysiwyg/pdfs/protocols/12888.pdf.

Padgett-Stewart, T. M. et al. (2015). An eDNA assay for river otter detection: A tool for surveying a semi-aquatic mammal. Conservation Genetics Resources, 8(1), 5-7. doi:10.1007/s12686-015- 0511x

Stoeckle, M. Y., Mishu, M. D., & Charlop-Powers, Z. (2018). GoFish: A Streamlined Environmental DNA Presence/Absence Assay for Marine Vertebrates. Retrieved from https://www.biorxiv.org/content/early/2018/05/25/331322

Stoeckle, M. Y., Soboleva, L., & Charlop-Powers, Z. (2017). Aquatic environmental DNA detects seasonal fish abundance and habitat preference in an urban estuary. Plos One, 12(4). doi:10.1371/journal.pone.0175186

Thomsen, P. F., & Willerslev, E. (2015). Environmental DNA – An emerging tool in conservation for monitoring past and present biodiversity. Biological Conservation, 183, 4-18. doi:10.1016/j.biocon.2014.11.019

Ushio, M. et al. (2017). Environmental DNA enables detection of terrestrial mammals from forest pond  water. Molecular Ecology Resources, 17(6). doi:10.1111/1755-0998.12690

Veldhoen, N. et al. (2016). Implementation of Novel Design Features for qPCR-Based eDNA Assessment. Plos One, 11(11). doi:10.1371/journal.pone.0164907

Weltz, K. et al. (2017). Application of environmental DNA to detect an endangered marine skate species in the wild. PLoS ONE, 12(6), e0178124. http://doi.org/10.1371/journal.pone.0178124


I would like to thank my mentor, Dr. Mark Stoeckle at The Rockefeller University’s Program for the Human Environment, for his help in teaching me how to execute lab procedures properly, working alongside me, expanding my knowledge of the field, and guiding my research to become a professional, well-structured study. Also, I would like to thank all state, county, and town representatives who I was in contact with and staff at the Prospect Park Zoo and Greenwich Audubon, who guided me in finding water sample collection sites. I would also like to thank my teachers, Mrs. Stephanie Greenwald, Mrs. Megan Salomone, and Dr. Caroline Matthew, and my family for their dedication to my research and helping me immensely along the way.