Sensitivity analyses for simulating pesticide impacts on honey bee colonies.

We employ Monte Carlo simulation and sensitivity analysis techniques to describe the population dynamics of pesticide exposure to a honey bee colony using the VarroaPop+Pesticide model. Simulations are performed of hive population trajectories with and without pesticide exposure to determine the effects of weather, queen strength, foraging activity, colony resources, and Varroa populations on colony growth and survival. The daily resolution of the model allows us to conditionally identify sensitivity metrics. Simulations indicate queen strength and forager lifespan are consistent, critical inputs for colony dynamics in both the control and exposed conditions. Adult contact toxicity, application rate and nectar load become critical parameters for colony dynamics within exposed simulations. Daily sensitivity analysis also reveals that the relative importance of these parameters fluctuates throughout the simulation period according to the status of other inputs.

Estimating dermal contact soil exposure for amphibians.

Chemical exposure estimation through the dermal route is an underemphasized area of ecological risk assessment for terrestrial animals. Currently, there are efforts to create exposure models to estimate doses from this pathway for use in ecological risk assessment. One significant limitation has been insufficient published data to characterize exposure and to support the selection and parameterization of appropriate models, particularly for amphibians in terrestrial habitats. Recent publications measuring pesticide doses to terrestrial-phase amphibians have begun to rectify this situation. We collated and summarized available measurements of terrestrial amphibian dermal exposure to pesticides from 11 studies in which researchers measured tissue concentrations associated with known pesticide experimental application rates. This data set included tissue concentrations in 11 amphibian species and 14 different pesticides. We then compared the results of two screening exposure models that differed based on surface area scaling approaches as a function of body weight (one based on birds as surrogates for amphibians and another amphibian-specific) to the measured tissue residue concentrations. We define a false-negative rate for each screening model as the proportion of amphibians for which the predicted concentration is less than the observed concentration (i.e., underestimate), contrary to the intent of screening models, which are intended to have a bias for higher exposure concentrations. The screening model that uses birds as surrogates did not have any instances where estimated expected avian doses were less than measured amphibian body burdens. When using the amphibian-specific exposure model that corrected for differences between avian and amphibian surface area, measured concentrations were greater than model estimates for 11.3% of the 1158 comparisons. The database of measured pesticide concentrations in terrestrial amphibians is provided for use in calculating bioconcentration factors and for future amphibian dermal exposure model development. Integr Environ Assess Manag 2023;19:9-16. © 2022 SETAC. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

A retrospective analysis of honey bee (Apis mellifera) pesticide toxicity data.

Current USEPA ecological risk assessments for pesticide registration include a determination of potential risks to bees. Toxicity data are submitted to support these assessments and the USEPA maintains a large database containing acute and chronic toxicity data on adult and larval honey bees (Apis mellifera), which USEPA considers a surrogate for Apis and non-Apis bees. We compared these toxicity data to explore possible trends. This analysis indicated a significant correlation between acute contact and oral median lethal dose (LD50) values for adult honey bees (ρ = 0.74, p <0.0001). Using default EPA modeling assumptions, where exposure for an individual bee is roughly 12x lower through contact than through ingestion, the analysis indicates that the oral LD50 is similarly if not more protective of the contact LD50 for the majority of pesticides and modes of action evaluated. The analysis also provided evidence that compounds with a lower acute toxicity for adults through contact and oral exposure pathways may still be acutely toxic for larvae. The acute toxicity of herbicides and fungicides was higher for larvae relative to oral and contact toxicity for adult honey bees for the same compounds and the no observed adverse effect level (NOAEL) from chronic toxicity studies were lower for larvae relative to adults, indicating increased sensitivity of larvae. When comparing 8-day LD50 values between single dose larval acute studies to those derived from repeat dose 22-day larval chronic toxicity studies, the LD50 values derived from chronic studies were significantly lower than those from acute toxicity tests (Z = -37, p = 0.03).

A spatially explicit model for estimating risks of pesticide exposure to bird populations.

Pesticides are used widely in agriculture and have the potential to affect non-target organisms, including birds. We developed an integrated modeling system to allow for spatially-explicit evaluation of potential impacts to bird populations following exposures to pesticides. Our novel methodology builds upon three existing models: the Terrestrial Investigation Model (TIM), the Markov Chain Nest Productivity Model (MCnest), and HexSim to simulate population dynamics. We parameterized the integrated modeling system using information required under the Federal Insecticide, Fungicide, and Rodenticide Act, together with species habitat and life history data available from the scientific literature as well as landcover data representing agricultural areas and species habitat. Our case study of the federally threatened California Gnatcatcher (Polioptila californica) illustrates how the integrated modeling system can estimate the population-scale consequences of pesticide applications. We simulated impacts from two insecticides applied to wheat: one causing mortality (survival stressor), and the other causing reproductive failure (reproductive stressor). We observed declines in simulated gnatcatcher abundance and changes in the species' distribution following applications of each pesticide; however, the impacts of the two pesticides were different. Our methodology attempts to strike a balance between biological realism and model complexity and should be applicable to a wide array of species, systems, and stressors.

Predicting impacts of chemicals from organisms to ecosystem service delivery: A case study of endocrine disruptor effects on trout.

We demonstrate how mechanistic modeling can be used to predict whether and how biological responses to chemicals at (sub)organismal levels in model species (i.e., what we typically measure) translate into impacts on ecosystem service delivery (i.e., what we care about). We consider a hypothetical case study of two species of trout, brown trout (Salmo trutta; BT) and greenback cutthroat trout (Oncorhynchus clarkii stomias; GCT). These hypothetical populations live in a high-altitude river system and are exposed to human-derived estrogen (17α­ethinyl estradiol, EE2), which is the bioactive estrogen in many contraceptives. We use the individual-based model inSTREAM to explore how seasonally varying concentrations of EE2 could influence male spawning and sperm quality. Resulting impacts on trout recruitment and the consequences of such for anglers and for the continued viability of populations of GCT (the state fish of Colorado) are explored. inSTREAM incorporates seasonally varying river flow and temperature, fishing pressure, the influence of EE2 on species-specific demography, and inter-specific competition. The model facilitates quantitative exploration of the relative importance of endocrine disruption and inter-species competition on trout population dynamics. Simulations predicted constant EE2 loading to have more impacts on GCT than BT. However, increasing removal of BT by anglers can enhance the persistence of GCT and offset some of the negative effects of EE2. We demonstrate how models that quantitatively link impacts of chemicals and other stressors on individual survival, growth, and reproduction to consequences for populations and ecosystem service delivery, can be coupled with ecosystem service valuation. The approach facilitates interpretation of toxicity data in an ecological context and gives beneficiaries of ecosystem services a more explicit role in management decisions. Although challenges remain, this type of approach may be particularly helpful for site-specific risk assessments and those in which tradeoffs and synergies among ecosystem services need to be considered.

A framework for linking population model development with ecological risk assessment objectives.

The value of models that link organism-level impacts to the responses of a population in ecological risk assessments (ERAs) has been demonstrated extensively over the past few decades. There is little debate about the utility of these models to translate multiple organism-level endpoints into a holistic interpretation of effect to the population; however, there continues to be a struggle for actual application of these models as a common practice in ERA. Although general frameworks for developing models for ERA have been proposed, there is limited guidance on when models should be used, in what form, and how to interpret model output to inform the risk manager's decision. We propose a framework for developing and applying population models in regulatory decision making that focuses on trade-offs of generality, realism, and precision for both ERAs and models. We approach the framework development from the perspective of regulators aimed at defining the needs of specific models commensurate with the assessment objective. We explore why models are not widely used by comparing their requirements and limitations with the needs of regulators. Using a series of case studies under specific regulatory frameworks, we classify ERA objectives by trade-offs of generality, realism, and precision and demonstrate how the output of population models developed with these same trade-offs informs the ERA objective. We examine attributes for both assessments and models that aid in the discussion of these trade-offs. The proposed framework will assist risk assessors and managers to identify models of appropriate complexity and to understand the utility and limitations of a model's output and associated uncertainty in the context of their assessment goals. Integr Environ Assess Manag 2018;14:369-380. Published 2017. This article is a US Government work and is in the public domain in the USA.

Mechanistic modeling of insecticide risks to breeding birds in North American agroecosystems.

Insecticide usage in the United States is ubiquitous in urban, suburban, and rural environments. There is accumulating evidence that insecticides adversely affect non-target wildlife species, including birds, causing mortality, reproductive impairment, and indirect effects through loss of prey base, and the type and magnitude of such effects differs by chemical class, or mode of action. In evaluating data for an insecticide registration application and for registration review, scientists at the United States Environmental Protection Agency (USEPA) assess the fate of the insecticide and the risk the insecticide poses to the environment and non-target wildlife. Current USEPA risk assessments for pesticides generally rely on endpoints from laboratory based toxicity studies focused on groups of individuals and do not directly assess population-level endpoints. In this paper, we present a mechanistic model, which allows risk assessors to estimate the effects of insecticide exposure on the survival and seasonal productivity of birds known to forage in agricultural fields during their breeding season. This model relies on individual-based toxicity data and translates effects into endpoints meaningful at the population level (i.e., magnitude of mortality and reproductive impairment). The model was created from two existing USEPA avian risk assessment models, the Terrestrial Investigation Model (TIM v.3.0) and the Markov Chain Nest Productivity model (MCnest). The integrated TIM/MCnest model was used to assess the relative risk of 12 insecticides applied via aerial spray to control corn pests on a suite of 31 avian species known to forage in cornfields in agroecosystems of the Midwest, USA. We found extensive differences in risk to birds among insecticides, with chlorpyrifos and malathion (organophosphates) generally posing the greatest risk, and bifenthrin and λ-cyhalothrin (pyrethroids) posing the least risk. Comparative sensitivity analysis across the 31 species showed that ecological trait parameters related to the timing of breeding and reproductive output per nest attempt offered the greatest explanatory power for predicting the magnitude of risk. An important advantage of TIM/MCnest is that it allows risk assessors to rationally combine both acute (lethal) and chronic (reproductive) effects into a single unified measure of risk.

Weight of evidence evaluation of a network of adverse outcome pathways linking activation of the nicotinic acetylcholine receptor in honey bees to colony death.

Ongoing honey bee (Apis mellifera) colony losses are of significant international concern because of the essential role these insects play in pollinating crops. Both chemical and non-chemical stressors have been implicated as possible contributors to colony failure; however, the potential role(s) of commonly-used neonicotinoid insecticides has emerged as particularly concerning. Neonicotinoids act on the nicotinic acetylcholine receptors (nAChRs) in the central nervous system to eliminate pest insects. However, mounting evidence indicates that neonicotinoids also may adversely affect beneficial pollinators, such as the honey bee, via impairments on learning and memory, and ultimately foraging success. The specific mechanisms linking activation of the nAChR to adverse effects on learning and memory are uncertain. Additionally, clear connections between observed impacts on individual bees and colony level effects are lacking. The objective of this review was to develop adverse outcome pathways (AOPs) as a means to evaluate the biological plausibility and empirical evidence supporting (or refuting) the linkage between activation of the physiological target site, the nAChR, and colony level consequences. Potential for exposure was not a consideration in AOP development and therefore this effort should not be considered a risk assessment. Nonetheless, development of the AOPs described herein has led to the identification of research gaps which, for example, may be of high priority in understanding how perturbation of pathways involved in neurotransmission can adversely affect normal colony functions, causing colony instability and subsequent bee population failure. A putative AOP network was developed, laying the foundation for further insights as to the role of combined chemical and non-chemical stressors in impacting bee populations. Insights gained from the AOP network assembly, which more realistically represents multi-stressor impacts on honey bee colonies, are promising toward understanding common sensitive nodes in key biological pathways and identifying where mitigation strategies may be focused to reduce colony losses.

A framework for predicting impacts on ecosystem services from (sub)organismal responses to chemicals.

Protection of ecosystem services is increasingly emphasized as a risk-assessment goal, but there are wide gaps between current ecological risk-assessment endpoints and potential effects on services provided by ecosystems. The authors present a framework that links common ecotoxicological endpoints to chemical impacts on populations and communities and the ecosystem services that they provide. This framework builds on considerable advances in mechanistic effects models designed to span multiple levels of biological organization and account for various types of biological interactions and feedbacks. For illustration, the authors introduce 2 case studies that employ well-developed and validated mechanistic effects models: the inSTREAM individual-based model for fish populations and the AQUATOX ecosystem model. They also show how dynamic energy budget theory can provide a common currency for interpreting organism-level toxicity. They suggest that a framework based on mechanistic models that predict impacts on ecosystem services resulting from chemical exposure, combined with economic valuation, can provide a useful approach for informing environmental management. The authors highlight the potential benefits of using this framework as well as the challenges that will need to be addressed in future work. Environ Toxicol Chem 2017;36:845-859. © 2017 SETAC.

Assessing population-level effects of zinc exposure to brown trout (Salmo trutta) in the Arkansas River at Leadville, Colorado.

We assessed population-level risk to upper Arkansas River brown trout (Salmo trutta L.) due to juvenile exposure to Zn. During spring, individuals in the sensitive young-of-the-year life stage are exposed to elevated Zn concentrations from acid mine drainage. We built and used a simple life-history population model for the risk assessment, with survival and fecundity parameter values drawn from published data on brown trout populations located in the United States and Europe. From experimental data, we derived a toxicity model to predict mortality in brown trout fry after chronic exposure to Zn. We tested sensitivity of risk estimates to uncertainties in the life-history parameters. We reached 5 conclusions. First, population projections are highly uncertain. A wide range of estimates for brown trout population growth is consistent with the scientific literature. The low end of this range corresponds to an unsustainable population, a physically unrealistic condition due to combining minimum parameter values from several studies. The upper end of the range corresponds to an annual population growth rate of 281%. Second, excess mortality from Zn exposure is relatively more predictable. Using our exposure-response model for excess mortality to brown trout fry due to Zn exposure in the upper Arkansas River at the mouth of California Gulch in the years 2000 to 2005, we derived a mean estimate of 6.1% excess mortality (90% confidence interval = 1.6%-14.1%). Third, population projections are sensitive to all the parameters that contribute to the onset of reproduction. The weight of evidence suggests that young-of-the-year survival is most important; it is inconclusive about the ranking of other parameters. Fourth, population-level risk from Zn exposure is sensitive to young-of-the-year survival. If young-of-the-year survival exceeds 20% to 25%, then the marginal effect of excess juvenile mortality on population growth is low. The potential effect increases if young-of-the-year survival is less than 20%. Fifth, the effect of Zn on population growth is predictable despite high uncertainty in population projections. The estimate was insensitive to model uncertainties. This work could be useful to ecological risk assessors and managers interested in using population-level endpoints in other risk assessments.