Interactions between arsenic and natural organic matter (NOM) are key limiting factors during the optimisation of drinking water treatment when significant amounts of both must be removed. This work uses Response Surface Methodology (RSM) to investigate how they interact during their simultaneous removal by iron chloride coagulation, using humic acid (HA) as a model NOM substance. Using a three factor Box-Behnken experimental design, As and HA removals were modelled, as well as a combined removal response. ANOVA results showed the significance of the coagulant dose for all three responses. At high initial arsenic concentrations (200 p4/1), As removal was significantly hindered by the presence of HA. In contrast, the HA removal response was found to be largely independent of the initial As concentration, with the optimum coagulant dose increasing at increasing HA concentrations. The combined response was similar to the HA removal response, and the interactions evident are most interesting in terms of optimising treatment processes during the preparation of drinking water, highlighting the importance of utilizing RSM for such investigations. The combined response model was successfully validated with two different groundwaters used for drinking water supply in the Republic of Serbia, showing excellent agreement under similar experimental conditions. (C) 2016 Elsevier B.V. All rights reserved.