Kelps have an extensive distribution in Arctic coastal waters. However, quantifying their role in the Arctic food web and carbon cycle is challenged by the scarcity of documented geographical distribution, standing stocks and production. Here we present a framework based on an empirical function to predict the potential kelp distribution and their summer biomass as a function of seafloor irradiance and bathymetry. Predictions of biomass were limited to the lower-depth, light-limited range of the kelp distribution, where the fit of the empirical function was significant (from the depth of maximum biomass to the deepest kelp extent). The model was developed and tested in Kongsfjorden, Svalbard, and applied in six additional fjords in the Arctic. The predicted potential kelp biomass in the fjords ranged from 0.6 to 4.7 kg WW m−2 and was in good agreement with published values. The resulting kelp standing stock ranged from 0.4 to 300 Gg DW, corresponding to 0.2–109 Gg C. These potential estimates account for light limitation, but do not consider substrata or other factors limiting the kelp distribution area. We identified fjord-specific dependencies between predicted standing stocks and seafloor irradiance and between seafloor irradiance and its drivers (surface irradiance and water column light attenuation) but found no significant change between 2004 and 2022. Our framework provides a baseline for estimating potential kelp biomass from seafloor irradiance, which is expected to change with increasing sediment runoff causing coastal darkening.