Bulk transfer coefficients estimated as a function of atmospheric stability and sea-ice lead width are combined with Arctic meteorological observations and ice thickness data to calculate the sensitivity of turbulent flux estimates to changes in lead width, wind speed, air temperature, and ice thickness for a high-concentration ice peck. These results are considered in terms of bulk transfer parametrizations that use a fixed transfer coefficient or that address atmospheric stability only. On the basis of the fetch-sensitive parametrizations considered here, differences in lead width for widths up to about 200 m can exert a substantial influence on sensible heat transfer coefficients and heat flux from leads under typical Arctic conditions. Fluxes from an open water lead decrease by 34% if fetch increases from 10 m to 100 m. This effect is greatest for open water leads, decreases considerably as leads refreeze, and is negligible for ice thicker than about 0.3 m. If open or newly refrozen leads make up 2% of the ice cover, then an increase in mean fetch from 10 m to 100 m yields a decrease of about 2 W m-2 in areally averaged flux from the ice pack. Calculations using observed and theoretical lead width distributions suggest that paramterizing lead widths in a sea ice model can be done effectively using a single, representative lead width rather than requiring a full distribution of widths. When coupled to the lower atmospheric boundary layer using a bulk similarity theory model, this sensitivity of heat transfer to fetch results in substantially higher near-surface air temperatures over narrow leads, with equilibrium air temperatures decreasing by about 50% as fetch increases from 10 to 100 m. ¿ American Geophysical Union 1995