Environmental pollutants are health hazards and are typically transported during runoff events. Monitoring the loadings of these pollutants with auto-samplers require precise trigger thresholds to effectively account for total maximum daily loads (TMDLs) entering natural water bodies. Traditionally, auto-samplers are triggered by delaying the start of sampling until pollutant wave is present during rainfall event. The rainfall-related triggers are typically limited to small watersheds, where lag and travel times are consistent and predictable. However, in large and more complex watersheds, flow or stage is typically used either by a set threshold on change in instantaneous flow rate or water level. Generally, trigger thresholds are difficult to establish due to seasonal fluctuations in stream flow and variations in rainfall. This study investigated dual parameter trigger based on instantaneous change and variance from a moving average for flow and stage. Nineteen auto-samplers, installed within six watersheds of varying land uses in City of Kissimmee, FL, were evaluated over 3-year period. The results suggested that using 20- to 30-min moving average of 5-min sampling interval for both parameters was sufficient to detect pollutant waves with minimal false triggers. Also, change from average flow rate (∆Qave) and a percent change from average stage (∆Have%) were found to the preferred parameters. The ∆Have% values ranging from -0.012 to 0.20% and ∆Qave ranging from 0.014 to 0.850 m(3)/s were found to give effective results for all stations in the study area. It was also observed that these trigger thresholds varied with land use, stream flow condition, and auto-sampler locations within the watershed.