Excess nitrogen in storm runoff in agricultural regions contributes to poor water quality in coastal and marine environments, including the Great Barrier Reef. Constructed wetlands are one mechanism by which land holders and governments in tropical north Queensland are combating poor water quality. Wetlands are known to be effective at promoting nitrogen amelioration – reducing the nitrogen concentration of outflows relative to their inflows. Mathematical models o f n itrogen p rocesses i n wetlands a re u seful f or exploring the conditions under which wetlands are most effective in order to guide wetland design and management. Understanding the relative importance of, and model sensitivity to, the various biogeochemical processes (within the wetland) and the boundary conditions (influxes to the wetland) is critical to target monitoring and further research where it will best reduce uncertainty and improve decision making. This paper presents a sensitivity analysis of a single-box compartment model for nitrogen within a wetland in tropical north Queensland. Five coupled ordinary differential equations describe the masses of nitrate (NO3−N), ammonium (NH4−N), dissolved organic nitrogen (DON), particulate organic nitrogen (PON), and suspended sediment (SS) within the wetland and includes terms for the release and transformation of the nitrogen forms within the wetland. The sensitivity analysis includes an exploration of 21 model parameters, namely the transfer rate constants and half saturation constants, the temperature rate factor, and the nitrogen concentrations in rain, through both a vary one-at-a-time (OAT) analysis and an all-at-the-same-time analysis. The sensitivity of the model to the boundary conditions, that is the mass influxes of the various nitrogen forms and sediment entering the wetland, was explored using a OAT analysis applying a fixed p roportional variation to the mass influx time series for each of NO3−N, NH4−N, DON, PON, and S S. The sensitivity was measured using the metrics of cumulative removal of each of total nitrogen (TN), the different nitrogen forms (NO3−N, NH4−N, DON, PON), and SS over a four month study period for a case study wetland. The all-at-the-same-time analyses demonstrated that while the degree of removal showed considerable variation across the parameter space, the majority (90%) of scenarios had net TN removal, with all scenarios showing net NO3−N and net PON removal. DON was generated within the wetland under all scenarios, and ammonium was generated in 93% of cases. Understanding the conditions under which net TN generation could occur will be essential for effective management of treatment wetlands. TN removal was PON driven, and hence very sensitive to those parameters that drive PON removal, namely particle size and density. While the absence of re-suspension processes will lead to an over-estimation of PON removal, high PON influxes indicate that settling ponds to reduce PON inflow may be a useful st rategy to improve wetland performance. Considering the internal processes, after settling TN was most sensitive to the ammonium release rate resulting from the mineralisation process. Ammonium was more sensitive to sediment release rates and plant uptake processes over the nitrification p rocesses, m ost l ikely d ue t o t he anaerobic conditions that prevailed within the wetland. Sensitivity (within the top 10) to the temperature rate factor was also observed for all nitrogen forms (and SS); further investigation of temperature impacts is therefore warranted, particularly in wetlands threatened by climate change. TN removal was PON driven, and hence very sensitive to those parameters that drive PON removal, namely particle size and density. While the absence of re-suspension processes will lead to an over-estimation of PON removal, high PON influxes indicate that settling ponds to reduce PON inflow may be a useful st rategy to improve wetland performance. Considering the internal processes, after settling TN was most sensitive to the ammonium release rate resulting from the mineralisation process. Ammonium was more sensitive to sediment release rates and plant uptake processes over the nitrification p rocesses, m ost l ikely d ue t o t he anaerobic conditions that prevailed within the wetland. Sensitivity (within the top 10) to the temperature rate factor was also observed for all nitrogen forms (and SS); further investigation of temperature impacts is therefore warranted, particularly in wetlands threatened by climate change.