There has been a significant increase in the number of published tidal-stream energy resource assessments in recent years due to the growing availability of open-source hydrodynamic models, and freely available data for model bathymetry (e.g. GEBCO_2014 and ETOPO) and boundary conditions (e.g. TPXO, FES, EOT). This study examines how the choice of bathymetry and tidal constituents affects the quantification of a tidal-stream energy resource, by conducting sensitivity tests for the Gulf of California. We find that the mean KPD (Kinetic Power Density) and annual mean power are significantly underestimated when using just GEBCO_2014 or ETOPO bathymetry data sources on their own. For the Midriff region, between San Lorenzo and San Esteban Islands (herein the San Lorenzo Passage), the annual mean power potential was estimated to be around 50 MW when using freely available bathymetry data, while the annual mean power increased to ~200 MW when using a bespoke dataset that was a combination of GEBCO and higher-resolution bathymetry provided by CICESE (The Centre for Scientific Research and Higher Education at Ensenada). Current speeds reduce from 2.4 m/s when using high-resolution to around 1.2 m/s and 0.8 m/s when using open source bathymetry products. Finally, we compared the estimated energy using tidal levelspredicted from 29 tidal constituents compared with simulations that included just the principal semi-diurnal lunar (M2) and solar (S2) constituents. The annual mean KPD reduced by almost 1/3rd in the San Lorenzo Passage, when just considering M2 and S2 constituents, suggesting that diurnal and higher order harmonic constituents are important for accurate resource assessments in this region.