Four side-by-side pilot-scale vertical flow (VF) constructed wetlands of different designs were evaluated for the removal of eight widely used emerging organic contaminants from municipal wastewater (i.e. ibuprofen, acetaminophen, diclofenac, tonalide, oxybenzone, triclosan, ethinylestradiol, bisphenol A). Three of the systems were free-draining, with one containing a gravel substrate (VGp), while the other two contained sand substrate (VS1p and VS2p). The fourth system had a saturated gravel substrate and active aeration supplied across the bottom of the bed (VAp). All beds were pulse-loaded on an hourly basis, except VS2p, which was pulse-loaded every 2h. Each system had a surface area of 6.2m2, received a hydraulic loading rate of 95mm/day and was planted with Phragmites australis. The beds received an organic loading rate of 7-16gTOC/m2d. The sand-based VF (VS1p) performed significantly better (p<0.05) than the gravel-based wetland (VGp) both in the removal of conventional water quality parameters (TSS, TOC, NH4-N) and studied emerging organic contaminants except for diclofenac (85±17% vs. 74±15% average emerging organic contaminant removal for VS1p and VGp, respectively). Although loading frequency (hourly vs. bi-hourly) was not observed to affect the removal efficiency of the cited conventional water quality parameters, significantly lower removal efficiencies were found for tonalide and bisphenol A for the VF wetland that received bi-hourly dosing (VS2p) (higher volume per pulse), probably due to the more reducing conditions observed in that system. However, diclofenac was the only contaminant showing an opposite trend to the rest of the compounds, achieving higher elimination rates in the wetlands that exhibited less-oxidizing conditions (VS2p and VGp). The use of active aeration in the saturated gravel bed (VAp) generally improved the treatment performance compared to the free-draining gravel bed (VGp) and achieved a similar performance to the free-draining sand-based VF wetlands (VS1p).