Ammonia (NH3) volatilization from fertilized turfgrass may represent a worst-case scenario: surface application without incorporation, especially if irrigation is not available. Although studied in field crop systems, field-scale losses of NH3 from turfgrass systems have received little study. Additionally, volatilization research with slow-release N sources is largely absent from the turfgrass literature. The objective of this study was to quantify NH3 volatilization from three N sources applied to warm-season turfgrass, using a field-scale mast system not yet evaluated in turfgrass settings. Treatments were applied to hybrid bermudagrass [Cynodon dactylon (L.) Pers. x Cynodon transvaalensis Burtt Davy] or zoysiagrass (Zoysia japonica Steud.) sod in October 2005, May 2006, and September 2006. Fertilizer treatments were urea (46-0-0), methylene urea (40-0-0), and polymer coated urea (41-0-0) applied at a N rate of 146 kg ha(-1). For 10 d NH3 was captured using a passive micrometeorological system of oxalic acid coated traps mounted on a rotating mast in the center of a 15 m diameter circular plot. In October 2005, NH3 volatilized from plots fertilized with urea was 11.7% of total N applied, which is significantly more volatilization than measured from plots receiving polymer coated urea. In May 2006 significantly more NH3 volatilized from plots fertilized with urea (11.7%) than from those fertilized with polymer-coated urea (3.6%). The 2006 studies had similar results, with significantly more NH3 volatilized from urea fertilized plots (average of 17.9%) than from plots fertilized with methylene urea (5.8%) or polymer-coated urea (4.4%). Of the three N sources, urea volatilized a greater proportion of applied N than the slow-release sources methylene urea and polymer coated urea.