Drinking water production needs to increasingly consider removal of background organic matter and trace micropollutants without increasing disinfection-by-product (DBP) formation potential. The presented data demonstrates the efficacy of both UV/H2O2 and UV/TiO2 in removing the pesticide metaldehyde to below drinking water compliance levels in both real and synthetic waters. This pesticide has proven to be unaffected by conventional water treatment processes such as granular activated carbon and is responsible for many of the water company compliance failures in the UK. The potential of UV/H2O2 is further demonstrated to offer an alternative approach for the removal of recalcitrant organic matter to ensure DBP compliance as long as extended UV doses of over 10,000 mJ cm(-2) are applied at the optimum peroxide dose of 8 mM. Alkalinity and UV dose have an impact on DBP formation: at low UV fluences, increased alkalinity reduced the DBP formation. The UV/TiO2 process was observed to be inhibited in the presence of alkalinity. Aggregation studies and comparison of the catalyst fractal dimension showed that the process inhibition is mainly due to aggregation. This restricts the surface area available for reactions, rather than changes in the catalyst properties or carbonate radical scavenging, which is often the reasoning attributed to photocatalysis inhibition. Hence, the presented results indicate that decreasing the catalyst aggregation is the key to apply photocatalysis as drinking water treatment. (C) 2016 Published by Elsevier B.V.