Debris covers glaciers worldwide and controls sub-debris melt rates by modifying energy transfer from the atmosphere to the ice. Physical properties like thermal conductivity (k) and surface roughness (z0) have been derived from limited local measurements, with models often relying on literature values from a few sites and studies. Accurate representation of these properties in energy-balance models is crucial for understanding climate-glacier interactions and predicting future behaviour of debris-covered glaciers. We studied these properties using established and modified approaches to derive k and z0 from field data at three locations on Pirámide Glacier in the central Chilean Andes. We compared existing methods and evaluated the modelled melt using these values. Our study reveals substantial inconsistencies between methods, leading to discrepancies between ice melt from energy-balance simulations and observed data, highlighting the impact of method choice on calculated ice melt. For energy-balance modelling, optimising k against measured ice melt appears a viable method to constrain melt simulations. Determining z0 is less critical due to its smaller impact on total ice melt, and profile aerodynamic method measurements, despite higher economic costs, are independent of ice melt calculations. The large, unexpected differences between existing methods indicate a substantial knowledge gap that the community should address. Values of k and z0 from field measurements do not work well when used in energy-balance models, suggesting that model values are bulk properties that do not necessarily correspond to field-derived values from theoretical approaches.