The heterogeneity of sea ice in the marginal ice zone (MIZ) causes multiscale secondary circulations that are challenging to model or observe. Their absent or inadequate representation in ocean-atmosphere interactions in climate models is partially responsible for the underestimation of Arctic sea ice loss. Observationally, such circulations obfuscate the interpretation of polar atmospheric chemistry measurements, among others. To address this open challenge, large-eddy simulations are conducted over real-world satellite-sensed sea ice maps, and idealizations of these maps that alter the ice pattern but conserve its fraction, showing that the ice fraction of a surface is not sufficient to predict surface heat flux. In a second suite of simulations, three other heterogeneity metrics (representing the surface fragmentation, tortuosity, and patch size variability) are then introduced; varying them among simulations suggests that surface-air interactions will be significantly impacted. Based on these results, we then show, using a multi-linear regression of map features, that MIZ surface parameterizations in Earth Systems Models can be improved if they account for these various features of the surface. Finally, the simulations also illustrate that the contribution of roughness changes to the resulting atmospheric circulations are minimal compared to thermal heterogeneity in the MIZ.

While yearly budgets of CO2 and evapotranspiration (ET) above forests can be readily obtained from eddy-covariance measurements, the quantification of their respective soil (respiration and evaporation) and canopy (photosynthesis and transpiration) components remains an elusive yet critical research objective. To this end, methods capable of reliably partitioning the measured ET and F_c fluxes into their respective soil and plant sources and sinks are highly valuable. In this work, we investigate four partitioning methods (two new, and two existing) that are based on analysis of conventional high frequency eddy-covariance (EC) data. The physical validity of the assumptions of all four methods, as well as their performance under different scenarios, are tested with the aid of large eddy simulations, which are used to replicate eddy-covariance field experiments. Our results indicate that canopies with large, exposed soil patches increase the mixing and correlation of scalars; this negatively impacts the performance of the partitioning methods, all of which require some degree of uncorrelatedness between CO2 and water vapor. In addition, best performance for all partitioning methods were found when all four flux components are non-negligible, and measurements are collected close to the canopy top. Methods relying on the water-use efficiency (W) perform better when W is known a priori, but are shown to be very sensitive to uncertainties in this input variable especially when canopy fluxes dominate. We conclude by showing how the correlation coefficient between CO2 and water vapor can be used to infer the reliability of different W parameterizations.