Using field data to quantify chemical impacts on wildlife population viability

Abstract

Abstract. Environmental pollution is an important driver of biodiversity loss. Yet, to date, the effects of chemical exposure on wildlife populations have been quantified for only a few species, mainly due to a lack of appropriate laboratory data to quantify chemical impacts on vital rates. In this study, we developed a method to quantify the effects of toxicant exposure on wildlife population persistence based on field monitoring data. We established field-based vital-rate-response functions for toxicants, using quantile regression to correct for the influences of confounding factors on the vital rates observed, and combined the response curves with population viability modelling. We then applied the method to quantify the impact of DDE on three bird species: the White-tailed Eagle, Bald Eagle, and Osprey. Population viability was expressed via five population extinction vulnerability metrics: population growth rate (r 1 ), critical patch size (CPS), minimum viable population size (MVP), probability of population extirpation (PE), and median time to population extirpation (MTE). We found that past DDE exposure concentrations increased population extirpation vulnerabilities of all three bird species. For example, at DDE concentrations of 25 mg/kg wet mass of egg (the maximum historic exposure concentration reported in literature for the Osprey), r 1 became small (White-tailed Eagle and Osprey) or close to zero (Bald Eagle), the CPS increased up to almost the size of Connecticut (White-tailed Eagle and Osprey) or West Virginia (Bald Eagle), the MVP increased up to approximately 90 (White-tailed Eagle and Osprey) or 180 breeding pairs (Bald Eagle), the PE increased up to almost certain extirpation (Bald Eagle) or only slightly elevated levels (White-tailed Eagle and Osprey) and the MTE became within decades (Bald Eagle) or remained longer than a millennium (White-tailed Eagle and Osprey). Our study provides a method to derive species-specific field-based response curves of toxicant exposure, which can be used to assess population extinction vulnerabilities and obtain critical levels of toxicant exposure based on maximum permissible effect levels. This may help conservation managers to better design appropriate habitat restoration and population recovery measures, such as reducing toxicant levels, increasing the area of suitable habitat or reintroducing individuals.

Mean Species Abundance as a Measure of Ecotoxicological Risk

“…Fifth, we assumed that is proportional to . Although many studies report relative changes in intrinsic growth rate ( to be proportional to those in carrying capacity ; Hakoyama et al 2000; Nakamaru et al 2003; Hendriks et al 2005; Hilbers et al 2018), there are also studies arguing against such a relationship (e.g., Bell 1990; Underwood 2006). Sixth, the calculations also disregard ecologically relevant processes like interspecies interactions (e.g., competition), which may lead to an underestimation of the impact of chemical stressors on a community level (de Laender et al 2008).…”

What Is the Impact of Accidentally Transporting Terrestrial Alien Species? A New Life Cycle Impact Assessment Model

“…In quantile regression, the response can represent any part of its probability distribution . Quantile regression is especially powerful for filtering out confounding factors in noisy explanatory data that obscure the true response and has been successfully applied to link exposure–response curves, for example, for pollution. − If the response is limited by confounding factors, quantile regression based on one of the upper boundaries of its probability distribution (e.g., the 95th percentile) is expected to reflect the response’s relationship to the corresponding explanatory variable . Conversely, one of the lower boundaries (e.g., the 5th percentile) reflects a relevant relationship if confounding factors boost the response.…”

Overcoming Challenges of Incorporating Higher Tier Data in Ecological Risk Assessments and Risk Management of Pesticides in the United States: Findings and Recommendations from the 2017 Workshop on Regulation and Innovation in Agriculture

“…Increased computational power and agreement on probabilistic risk paradigms allowed regional scale assessments to emerge (Solomon et al 1996Purucker et al 2007). Recent models may incorporate more refined predictions of exposures and effects across landscapes (Schmolke et al 2010;Dixon 2012;Bartell et al 2013;Kohler and Triebskorn 2013;Van den Brink 2013;Wang 2013;Focks et al 2014;Topping et al 2015;Dohmen et al 2016;Hilbers et al 2018). These approaches allow prediction of population exposures to pesticides and resultant responses without having to conduct painstaking multiyear field assessments of population dynamics.…”