A comparative study of the modeled effects of atrazine on aquatic plant communities in midwestern streams

Abstract

Potential effects of atrazine on the nontarget aquatic plants characteristic of lower-order streams in the Midwestern United States were previously assessed using the Comprehensive Aquatic System Model (CASMATZ ). Another similar bioenergetics-based, mechanistic model, AQUATOX, was examined in the present study, with 3 objectives: 1) to develop an AQUATOX model simulation similar to the CASMATZ model reference simulation in describing temporal patterns of biomass production by modeled plant populations, 2) to examine the implications of the different approaches used by the models in deriving plant community-based levels of concern (LOCs) for atrazine, and 3) to determine the feasibility of implementing alternative ecological models to assess ecological impacts of atrazine on lower-order Midwestern streams. The results of the present comparative modeling study demonstrated that a similar reference simulation to that from the CASMATZ model could be developed using the AQUATOX model. It was also determined that development of LOCs and identification of streams with exposures in excess of the LOCs were feasible with the AQUATOX model. Compared with the CASMATZ model results, however, the AQUATOX model consistently produced higher estimates of LOCs and generated non-monotonic variations of atrazine effects with increasing exposures. The results of the present study suggest an opportunity for harmonizing the treatments of toxicity and toxicity parameter estimation in the CASMATZ and the AQUATOX models. Both models appear useful in characterizing the potential impacts of atrazine on nontarget aquatic plant populations in lower-order Midwestern streams. The present model comparison also suggests that, with appropriate parameterization, these process-based models can be used to assess the potential effects of other xenobiotics on stream ecosystems.

Modeling the effects of thiamethoxam on Midwestern farm ponds and emergent wetlands

“…The annual time scale is also consistent with previous uses of the CASM for assessing ecological risks for pesticides (Nair et al 2015;Bartell et al 2013). Although it is conceivable that longer-term food web and ecosystem effects might result from continued applications of the pesticide, the environmental chemistry of thiamethoxam suggests that it will degrade within a single year following application.…”

“…Population-specific lethal endpoints (e.g., LC50) for thiamethoxam were used to define corresponding exposure-response functions and compute corresponding toxic effect factors that adjusted modeled baseline mortality rates as a function of the daily exposure concentration (Nair et al 2015;Bartell et al 2013). Similarly, population-specific sublethal endpoints (e.g., EC50) were used to compute the toxic effect factors that adjusted bioenergetics parameters to produce the expected reduction in growth based on the exposure concentration.…”

“…The AQUA-TOX formulations for bioaccumulation and associated toxic effects were incorporated into the CASM GFP and CASM GWL to compare and contrast the 2 approaches for estimating toxic effects of thiamethoxam (Nair et al 2015;Park and Clough 2012). The AQUA-TOX formulations for bioaccumulation and associated toxic effects were incorporated into the CASM GFP and CASM GWL to compare and contrast the 2 approaches for estimating toxic effects of thiamethoxam (Nair et al 2015;Park and Clough 2012).…”

A weight‐of‐evidence approach for deriving a level of concern for atrazine that is protective of aquatic plant communities

“…The LOC using the up‐to‐date cosm database was 19.6 µg/L (Nair et al ). The weight‐of‐evidence score was 11 out of 15 (Table ) indicating that the LOC derived with CASM ATZ2 method has high confidence.…”

“…Each of the methods has strengths and weaknesses. Our evaluation determined that the CASM ATZ2 method (Bartell et al ; Nair et al ) has the highest reliability because it directly simulates effects to plant communities, can simulate time varying atrazine concentrations, and has performed well in the validation efforts conducted to date. The least reliable method is the Brock et al () method primarily because most of the input values near the range of possible LOCs for atrazine are of unacceptable quality.…”