Solar transformation of herbicides in prairie pothole wetlands: dynamics of indirect processes

Andrew J. McCabe and William A. Arnold, Department of Civil, Environmental, and Geo- Engineering

Prairie pothole wetlands of the Northern Great Plains are highly impaired systems resulting from agricultural encroachment, oil exploration, and climate change. The impacts range from habitat degradation and changes in the make-up of microbial communities to herbicide contamination and loss of floodwater storage.

William Arnold and Andrew McCabe’s USGS-funded work focuses on water quality and determining the capacity of prairie pothole wetlands to attenuate herbicides. In prairie pothole surface waters, photochemical transformation offers a unique attenuation pathway of herbicides because the concentration of two photo-sensitizers can be quite high: dissolved organic carbon ranges between 20 – 40 mg C/L and nitrate, in cases where wetlands receive agricultural runoff, reaches concentrations up to 2 mg NO3-/L. These sensitizers are significant because when exposed to light, they produce reactive species (radicals and excited-state molecules) that can react with contaminant herbicides. Where these reactive species occur in “high” concentrations (approx. 1 part per trillion), they significantly decrease the half-life of an herbicide, especially for herbicides that do not readily absorb sunlight themselves.

Because solar intensity and water chemistry vary daily and seasonally, steady-state concentrations of reactive species are highly dynamic. Arnold and McCabe note that half-lives for a single herbicide can vary by orders of magnitude depending on the water sample. This dynamic nature makes accurate prediction of photochemical half-lives difficult and necessitates an accurate model that estimates steady-state concentrations of reactive species. Models for prairie pothole wetlands based on alkalinity, pH, dissolved organic carbon concentration, nitrate and nitrite concentrations, and UV and visible light absorption will predict steady-state concentrations of four reactive species: hydroxyl radical, singlet oxygen, carbonate radical, and triplet excited-state dissolved organic matter.  

The steady-state concentrations of these reactive species are inherently linked. With the development of these models, we will be able to quantitatively predict how the half-life of a contaminant herbicide will change as a function of water chemistry. The ultimate goal is to know, in cases where herbicide contamination is a risk to human or environmental health, the length of time it will take for natural biogeochemical processes to eliminate the risk of exposure.

It is possible that this model could be used to enhance photochemical transformation of contaminant herbicides in prairie wetlands. It is, however, only in best-case scenarios that contaminants are completely phototransformed to non-toxic or biodegradable products. Photochemistry does not happen in isolation; there are many other process occurring simultaneously (e.g., sorption or dilution) that prevent phototransformation from occurring. Additionally for some contaminants, phototransformation does not change toxicity or toxicity is possibly enhanced.

Phototransformation is an important biogeochemical process, and must be understood within the context of ecotoxicology and other physical and biological processes. The model developed by this work contributes a piece to the collective understanding of environmental organic contaminant processing.


Andrew McCabe collecting a surface water sample from a temporary wetland at the Cottonwood Lakes Study Area near Jamestown, ND.