Fuzzy logic models are an effective tool for forecasting. However, few studies comparing different fuzzy logic models and their applications to river ice forecasting have been reported. This paper evaluates the application of two types of fuzzy logic models (a Qualitative Fuzzy Logic Model, QFLM and an Adaptive Neuro-fuzzy inference system, ANFIS) and an alternative model (Multiple Linear Regression, MLR) to predict the maximum water level during river ice breakup. The Athabasca River is the largest unregulated river in Alberta, Canada with ice jams frequently occurring in the vicinity of Fort McMurray. River ice breakup data for the Athabasca River at Fort McMurray, over the past 39 years (1977-2015), have been collected to facilitate the model comparisons. The results indicated that the QFLM can generate a qualitative evaluation and be treated as a pre-screening model for overall assessment of ice- caused flooding risk at breakup. As for quantitative prediction of deterministic values of maximum breakup water level, the fitting and predictive abilities of ANFIS are relatively better than those of MLR. In practice, both ANFIS and MLR can be used as forecast and backup tools, respectively. Further improvement of these models is still needed in terms of the selection of indicators and updating of datasets. These models lay the basis for effectively supporting spring breakup monitoring operations and emergency response to ice-related flooding.