Dragonfly Larvae Show Mercury Varies Widely Across Glen Canyon

The five places in Glen Canyon National Recreation Area where dragonfly larvae were sampled in 2018 are all connected by water: the San Juan and Colorado Rivers converge and flow into Lake Powell. Yet the mercury in the larvae ranged from 32 parts per billion, dry weight, at The Slough to 1252 at Slickhorn Canyon. The finding comes from the Dragonfly Mercury Project (DMP), described in an article by National Park Service (NPS) and U.S. Geological Survey (USGS) scientists with a University of Maine researcher.

Over 100 National Park Service units contributed to the project from 2009 to 2020 (NPS Air Resources Division).

Why dragonflies

Mercury is a harmful air pollutant that poses health risks to people and wildlife, and can affect reproduction and neurological function in humans. Most of the mercury people release today comes from coal combustion and gold mining. It can travel thousands of miles in wind, dust, rain and snow and settle in even remote places. Once it lands, natural processes convert it to methylmercury, which can enter cells and build up in the tissues of wildlife and people (bioaccumulation).

Sources and paths of mercury in the environment. Once in the food chain, mercury bioaccumulates and also biomagnifies, increasing in concentration with each step up the food chain (NPS Air Resources Division).

Dragonfly larvae work as biosentinels, or indicators, of mercury in aquatic food webs. They live underwater for months to years as predators, eating smaller insects and accumulating mercury from that food. They are also prey for fish, amphibians and birds, which passes mercury along a food web that may eventually pose health risks to people who eat fish. Sampling larvae therefore indicates the potential risk of mercury impairment in the wider ecosystem.

Juvenile dragonflies can spend up to nine years as predatory larvae, accumulating mercury as they feed (NPS/Brad Jorgensen).

The DMP engages citizen scientists, members of the public who volunteer to contribute to the scientific process, to collect larvae in national parks for mercury analysis. Since 2009 it has involved 121 park units and more than 5,000 citizen scientists who gave over 20,000 hours of service, and its data on mercury concentrations are publicly available to help resource managers and agencies protect human and wildlife health.

Sampling at Glen Canyon

Glen Canyon joined for the first time in 2018. Its biologists enlisted five youth organizations: Grand Canyon Youth Middle School Adventure, Page Boy Scout Troop 7911, Page Girl Scout Troop 7489, Page High School and the Southwest Conservation Corps Ancestral Lands Native Conservation Corps. A total of 53 students and 23 adult leaders took part and gave 231 hours to the project that year.

Between June and August, park staff and citizen scientists sampled five sites, chosen for their reliable larvae populations and wide geographical and altitudinal ranges: Lone Rock Beach, The Slough (next to the Colorado River), Bowns Canyon, Ribbon Canyon and Slickhorn Canyon (next to the San Juan River).

At each site the volunteers helped record physical structure, vegetation and geology, then swept D-frame dip nets through vegetation and sediments under the water. Each larva was handled with clean techniques to avoid contamination, bagged, measured to the nearest millimeter and identified to family with field cards from the project. Professional staff trained in Odonata, the order that includes dragonflies and damselflies, later validated the identifications. Family matters because families differ slightly in diet, habitat and physiology, but the researchers were able to model mercury variation statistically at the family level and compare across families. The samples were frozen and shipped to the USGS laboratory in Corvallis, Oregon, for total mercury analysis.

What the larvae showed

Each site falls into one of five categories of mercury risk: sub-impairment, low, moderate, high or severe. The categories were set partly by how many existing wildlife risk benchmarks the concentration exceeds; the more it exceeds, the higher the potential risk.

The park-wide average in 2018 was 241 ± 54 parts per billion, dry weight, which the article says is more than twice the project-wide average of 113 for that year. Across the project since 2009, larval mercury can vary up to 44-fold within a park and 135-fold among parks. Of the 457 sites sampled from 2009 to 2018, most show moderate to low risk, but 12 percent are at high or severe risk.

What it suggests

The authors point out that even though the same water may flow through all the Glen Canyon sites, their larvae differ markedly. That suggests site characteristics, such as plants, elevation, geology, hydrology and water chemistry, help convert mercury to toxic methylmercury at different rates. Air pollution sources like coal-fired power plants matter for deposition, but the landscape is also extremely important in how much mercury ends up in a habitat and, finally, in the food chain.

A site at higher risk could mean potential risk to people: if larvae at a site exceed an existing fish health benchmark, wildlife and people who eat fish from there could face elevated risk.

Growing the project

In 2018, park staff and citizen scientists collected larvae from 143 sites across 57 NPS units and National Natural Landmarks, bringing the number of participating parks to 102; the total reached 107 in 2019 and 121 after the 2020 season. The data now cover water bodies in 47 states, including Alaska and Hawaii. In 2018 a Steering Committee was formed with members of NPS, USGS, the U.S. Fish and Wildlife Service and the U.S. Environmental Protection Agency to plan expansion to new partners, landscapes and research questions.

Sources

• National Park Service, The Dragonfly Mercury Project: Citizen Science in Action at Glen Canyon National Recreation Area, by K. Ko, L. Pilkington, C. Flanagan Pritz, S. J. Nelson, C. Eagles-Smith, J. Willacker and C. Emery, Intermountain Park Science; rewritten in hubnx's own words.