Nov. 16, 2021
Oct. 29, 2021
Mercury pollution, which has increased due to anthropogenic activities, can spread far from its sources. Knowing how tropical forests solve this issue is important, as these ecosystems are valuable carbon sinks which keep pollutants out of circulation. However, there are many gaps in the understanding of mercury flux in tropical forests. A team of scientists used a plant canopy imager to study how canopy characteristics during different forest successional stages could be influencing mercury flux.
Mercury exists naturally in the atmosphere, and it is released by volcanic activity or due to degassing from water surfaces and soils. However, preindustrial levels of atmospheric mercury were low. Due to anthropogenic activities such as the burning of fossil fuels and smelting, the current levels are higher. Once released, mercury vapor can stay in the atmosphere for a year before it is deposited back to soils and water through precipitation. See Figure 1 for a diagrammatic representation of the mercury cycle.
When it is deposited, it can accumulate in soils until it is released due to disturbances such as forest logging or fires.
Figure 1: Biogeochemical Cycling of Mercury, UWEC. (Image credits:
Mercury can travel far from its source during its time as vapor, making it a concerning global pollutant. The threat of mercury is made clear by the fact that it has reached remote areas of the Brazilian Amazon; high concentrations of mercury have been found on the hair, and feathers of both humans and wildlife in the region. These concentrations have even been found in blood and muscle tissue.
Given the difference in concentration in the various layers of the forest soils, it is clear that mercury levels in tropical forests are not only due to natural sources.
Tropical forests capture a lot of mercury, along with other substances in the air, due to their larger leaf surface area. Both rainwash from the canopy and litterfall add mercury to the soil. In addition, mercury-laden rain deposits the mercury directly into the soil. Rain and rainwash transfer a quarter of the mercury that falls in the forest to the soils.
The amount of mercury in litterfall can vary by a factor of six among species. Differences in leaf area and lifespan, along with other leaf morphological features such as trichomes, epicuticular wax, and stomatal density, could be responsible. These features influence photosynthesis and the absorption of mercury into the trees.
There is not much that is known about the mercury biogeochemical cycle in the tropics. To get a better understanding, and to quantify mercury flux in the tropics, Brazilian scientists from marine and geosciences studied the Brazilian Atlantic Forest. These forests are a biodiversity hotspot and could be important for the mercury cycle.
A group of Brazilian scientists—Teixeira, Lacerda, and Silva-Filho—checked the mercury concentrations in litterfall in three successional stages with varying canopy closure. They also collected physicochemical and biological data to study the influence of forest heterogeneity and biophysical factors of the canopy on mercury flux. Micrometeorological data over two years informed them of the site microclimate.
The study sites were the evergreen mountain rainforests in the Serra da Mantiqueira, a mountain range in Itatiaia National Park, Brazil. Twenty-four litter traps were laid to collect litterfall from three successional stages:
Each litter trap of an area of 0.25 m2 was placed 70 cm above the ground, and there were eight in each successional stage. Litter was collected every fortnight and two collections were combined to get the monthly sample. The litter was divided into parts like leaves, twigs, flowers, fruits, etc. Each fraction was weighed and dried separately and then pulverized. The samples were then combined again, and mercury was extracted by an acid mixture and measured by Cold Vapour Atomic Absorption Spectrophotometry.
A complete weather station recorded rainfall, temperature, solar radiation, wind direction and speed, and relative humidity every 15 minutes for over two years.
The canopy cover measurement tool used to measure canopy closure, at twenty sites for each successional stage, was the CI-110 Plant Canopy Imager, manufactured by CID Bio-Science Inc. The scientists made the canopy cover measurement between 10 and 11 am on cloudless days.
The canopy cover measurement tool has a digital fish-eye camera mounted on a long probe. It takes hemispherical images of the canopy from the ground and stores the image on the device, which can be downloaded for later analysis. The Gap Fraction Method shows how much of the sky is visible through the canopy to allow canopy closure calculations.
The micro-weather station showed the scientists that there were four dry winter months and eight rainy summer months in both years. During the wet season, rain in a 15-minute interval exceeded 5 mm and was enough to create rainwash of mercury from the leaves.
Table 1: Canopy closure in 20 plots of each of the three successional stages, as measured by the CI-110 Digital Plant Canopy Imager, Teixeira, Lacerda, and Silva-Filho 2017. (Credits: https://doi.org/10.1016/j.chemosphere.2016.10.081)
The Plant Canopy Imager showed a significant difference in the canopy closure in the three forest types. The oldest MS forests had between 75-100% canopy cover, the LS had 67%, and the youngest ES had 42% canopy cover; see Table 1 for details. The oldest forests resembled pristine forest conditions, while the other two stages were characteristic of secondary growth.
Maximum production of litterfall (>0.6 Mg ha-1 month-1) occurred during dry and windy months and experienced two prior months of shorter days and colder weather. This occurred between September-November in 2009 and between August-October in 2010. The lowest litterfall (0.3 Mg ha-1 month-1) occurred between April and June; see Figure 1.
The LS plots produced the maximum litterfall of 7.3 Mg ha-1 year-1, followed by the oldest MS plots that shed 5.5 Mg ha-1 year-1, and the youngest ES plots produced only 4.8 Mg ha-1 year-1; see Figure 1. The litter production varied only between LS and ES, and LS and MS plots. The oldest and youngest forests produced similar amounts of litter.
The MS stages are the oldest plots but have less litterfall, as they have few highly productive pioneer species. Moreover, the microclimate under the closed canopy of MS forests is free of water stress due to constant humidity and temperature, so there is less leaf shedding.
Figure 2: “Litterfall production, Hg concentrations and fluxes along the studied period at the Itatiaia National Park, SE e Brazil,” Teixeira, Lacerda, and Silva-Filho 2017. (Image credits: https://doi.org/10.1016/j.chemosphere.2016.10.081)
The levels of mercury in the litterfall progressively decreased as the forest age increased. Thus, ES, LS, and MS litterfall had 74.1 ng g-1, 52.8 ng g-1, and 48.4 ng g-1, respectively. Mercury was significantly different only between the ES and the other two stages. The early successional stages have a larger leaf area and higher stomatal concentration, which lead to more photosynthesis, giving the trees a chance to absorb more mercury, which is reflected in the litterfall. About 70-80% of mercury in leaves could be entering through stomata.
Litterfall had the least mercury after long torrential rains since much of the pollutant is washed off the leaves during the heavy rainfall.
The scientists concluded that photosynthesis was the main process that controlled the entry of mercury into leaves. Also, there was a correlation between an increase in solar radiation and mercury in litterfall throughout the experiment. The growing season in the forest coincides with high solar radiation and rains at the end of summer. Stomatal activity increases during this time to facilitate photosynthesis, which also increases mercury intake by the leaves. This is the reason why high mercury was seen in litterfall during the growing season in ES plots. Old MS forests with closed canopies, on the other hand, get only diffuse light; therefore, they have less stomatal opening and mercury concentrations.
The average mercury concentration in litterfall was 57 ng g−1, and the mercury flux or accumulation in the soils through litterfall was 34.6 μg m−2 yr−1. The scientist found that as the concentration of mercury in litterfall increased, so did the mercury flux through litterfall. The amount of litterfall produced was also important. Leaves were the fraction of litterfall that were correlated with mercury flux.
The species composition in the successional stages also influences mercury flux. The species found in younger secondary forests grow faster, due to an open canopy and increased solar radiation. Hence, they have increased photosynthesis and can absorb more mercury. In younger forests, the mercury concentration is important in adding mercury to the soil. Because older forests grow slower and accumulate less mercury, litter production is the important way to add mercury to the soil. ES, LS, and MS showed mercury flux of 37.2, 37.7, and 26.9 μg m-2, respectively.
Mercury in litterfall was ten times higher in these forests than in temperate and boreal forests, but the lowest among the Brazilian rainforests. They are also lower than urban forests in Rio de Janeiro. The lower concentration of mercury in these forests is due to their distance from pollution sources and because the climate at higher altitudes affects the rate of photosynthesis.
In tropical forests, litterfall is the main route of mercury deposition to the soils. The rate of photosynthesis, which depends on solar radiation moderated by canopy and climate, determines the amount of mercury that enters the leaves and reaches the forest floor as litter. This study also shows that the successional stages or age of forests will influence mercury flux. Thus, the mercury flux is dynamic. Tropical forest mercury sequestration is ten times greater than in temperate forests, so any consideration of land-use changes should include assessment of these important characteristics.
Science Writer, CID Bio-Science
Ph.D. Ecology and Environmental Science, B.Sc Agriculture
Feature image courtesy of Dimitry Novikov
Mercury in the Environment and Water Supply. Cycling. (n.d.). https://people.uwec.edu/piercech/hg/mercury_water/cycling.htm.
Selin, N. E. (2009). Global Biogeochemical Cycling of Mercury: A Review. Annual Review of Environment and Resources, 34(1), 43–63. https://doi.org/10.1146/annurev.environ.051308.084314
Teixeira, D. C., Lacerda, L. D., & Silva-Filho, E. V. (2017). Mercury sequestration by rainforests: The influence of microclimate and different successional stages. Chemosphere, 168, 1186–1193. https://doi.org/10.1016/j.chemosphere.2016.10.081
1-360-833-8835 ext. 217