629 Environmental Toxicology and Chemistry, Vol. 28, No. 3, pp. 629–636, 2009  2009 SETAC Printed in the USA 0730-7268/09 $12.00 + .00 COMBINED CHEMICAL (FLUORANTHENE) AND DROUGHT EFFECTS ON LUMBRICUS RUBELLUS DEMONSTRATE THE APPLICABILITY OF THE INDEPENDENT ACTION MODEL FOR MULTIPLE STRESSOR ASSESSMENT SARA M. LONG,† FREDRIK REICHENBERG,‡ LINDSAY J. LISTER,† PETER K. HANKARD,† JOANNA TOWNSEND,† PHILIPP MAYER,‡ JULIAN WRIGHT,† MARTIN HOLMSTRUP,§ CLAUS SVENDSEN,*† and DAVID J. SPURGEON† †Centre for Ecology and Hydrology, Monks Wood, Abbots Ripton, Huntingdon, Cambridgeshire PE28 2LS, United Kingdom ‡National Environmental Research Institute, University of Aarhus, P.O. Box 358, 4000 Roskilde, Denmark §National Environmental Research Institute, University of Aarhus, Department of Terrestrial Ecology, P.O. Box 314, Vejlsøvej 25, DK-8600 Silkeborg, Denmark ( Received 23 April 2008; Accepted 28 September 2008) Abstract—The combined effect of a chemical (fluoranthene) and a nonchemical stress (reduced soil moisture content) to the widely distributed earthworm Lumbricus rubellus were investigated in a laboratory study. Neither fluoranthene (up to 500 g/g) nor low soil moisture (15% below optimal) had a significant effect on the survival of the exposed worms, but a significant effect on reproduction (cocoon production rate) was found for both stressors ( p  0.001 in both cases). The response of cocoon production to each stressor could be well described by a logistic model; this suggested that the joint effects may be applicable to description using the independent action (IA) model that is widely used in pharmacology and chemical mixture risk assessment [1]. Fitting of the IA model provided a good description of the combined stressor data (accounting for 53.7% of total variation) and was the most parsimonious model describing joint effect (i.e., the description of the data was not improved by addition of further parameters accounting for synergism or antagonism). Thus, the independent action of the two responses was further supported by measurement of internal fluoranthene exposure. The chemical activity of fluoranthene in worm tissue was correlated only with soil fluoranthene concentration and not with soil moisture content. Taken together these results suggest that the IA model can help interpret the joint effects of chemical and nonchemical stressors. Such analyses should, however, be done with caution since the literature data set suggests that there may be cases where interactions between stressors result in joint effects that differ significantly from IA predictions. Keywords—Drought Polycyclic aromatic hydrocarbons Multiple stressors Independent action Chemical activity INTRODUCTION Processes undertaken in soil are essential for global bio- geochemical cycles. Soils act as a sink for carbon; they supply essential nutrients to support food and biomass production; store, buffer, filter, and transform pollutants; and filter and regulate the water supply. It is not the physical constituents alone but rather the interaction between physical properties and soil biological communities that facilitate the delivery of ecosystem services [2,3]. Ample evidence suggests that human disturbance impacts on soil communities, thereby affecting soil functioning [4] and food web stability [5]. Climate change and the presence of toxic pollutants currently represent two of the most serious threats to soil function. Predicting the field effects of natural and anthropogenic stressors such as drought conditions or chemicals on soil spe- cies is frequently complicated by the fact that soil ecosystems may be simultaneously subjected to multiple pressures. Thus, effects of changing land management practice may be overlaid by pollution, all this occurring against the background of glob- al climatic change. This has encouraged research into the na- ture of combined effects [6–9]. Results of these studies have illustrated that interactions can occur between chemical and other stressors, such as drought, that can increase chemical toxicity to soil biological communities. * To whom correspondence may be addressed (csv@ceh.ac.uk). Published on the Web 11/3/2008. Clearly, predicting the impact of such multiple stressors scenarios on the ecology (e.g., life history traits) of even an individual soil species is a challenging task and one that clearly needs a sound theoretical framework from which effect pre- dictions can be made. The basis for such a framework can be found in the classical mathematics of probability theory, spe- cifically the calculation of joint probability of independent events. This theory was related to chemical mixtures by Bliss [1], calculating the proportion affected by joint exposure through combining the risk of effects from individual chem- icals derived from their dose–response curves. Here we sought to test whether this model, originally developed for chemical mixtures, was able to predict the combined effects of a chem- ical–environmental stressor combination to the widely distrib- uted earthworm species Lumbricus rubellus and so could be useful for future combined effect assessment. The two stressors investigated in combination were the polycyclic aromatic hy- drocarbon (PAH) fluoranthene and the nonchemical stressor of reduced soil moisture content. These exposure conditions were chosen as a realistic scenario since an increase in the frequency and severity of summer droughts is one of the most frequently predicated consequences of future climate change [10], while extensive soil surveys have shown the widespread contamination of soils by PAHs [11]. The model selected to predict the joint effect of the two stressors was the independent action (IA) model [1]. This mod- el has already been widely applied to predict the effects of mixtures of dissimilarly acting chemicals in both aquatic [12–