An experimental water treatment plant was established to verify the effectiveness of constructed wetlands to improve water quality in the Venice Lagoon watershed. The wetland comprised three different subsystems, ranging from a riparian swamp to a marsh ecosystem. As a first step, monitoring was conducted over three years to evaluate the efficacy and efficiency of the system. Here, we report an analysis of the water budget resulting from routinely collected hydraulic and meteorological data. We used independent estimates of the water-budget terms, rather than budgetary residual estimation, because we wanted to estimate the water-budget error. Surface-water inflow, surface-water outflow, and direct precipitation were measured. Daily potential evapotranspiration values were estimated using the FAO Penman Monteith equation; runoff was estimated using the USDA Soil Conservation Service curve number model; indirect precipitation that flows toward the wetland via subsurface flow was estimated from the soils field capacity; and seepage was estimated using Darcy's law. The objectives of the research were to establish the best way to develop a water budget useful for application and design purposes, to determine the term that most influences the water-budget error, and to perform a sensitivity analysis on the parameters affecting this term. Surface flow dominated the wetland system, precipitation and evapotranspiration contributed about 10% to the water budget, and changes in water storage and the seepage flow contributed less than 5% to the water budget. The seepage flow term had the highest uncertainty, and the frequency of ground-water level measurements had the greatest impact on the water-budget error (ranging from 10.0%–28.6%). Therefore, in a free-water surface wetland with a shallow ground-water system, the main effort in field measurement should be to ensure a measurement frequency of less than five days.