The effect of spatial sampling on magnetic field modeling and helicity computation
Nonlinear force-free (NLFF) modeling is regularly used in order to indirectly infer the 3D geometry of the coronal magnetic field, not accessible on a regular basis by means of direct measurements otherwise. We study the effect of binning in time series NLFF modeling of individual active regions (ARs) in order to quantify the effect of a different underlying spatial sampling on the quality of modeling as well as on the derived physical parameters. We apply an optimization method to sequences of SDO/HMI vector magnetogram data at three different plate scales for three solar ARs to obtain nine NLFF model time series. From the NLFF models, we deduce active-region magnetic fluxes, electric currents, magnetic energies and relative helicities, and analyze those with respect to the underlying spatial sampling. We calculate various metrics to quantify the quality of the derived NLFF models and apply a Helmholtz decomposition to characterize solenoidal errors. At a given spatial sampling, the quality of NLFF modeling is different for different ARs, as well as varies along of the individual model time series. For a given AR, modeling at a given spatial sampling is not necessarily of superior quality compared to that performed with a different plate scale. Generally, the NLFF model quality tends to be higher for larger pixel sizes with the solenoidal quality being the ultimate cause for systematic variations in model-deduced physical quantities. Optimization-based modeling based on binned SDO/HMI vector data delivers magnetic energies and helicity estimates different by $\lesssim$30\%, given that concise checks ensure the physical plausibility and high solenoidal quality of the tested model. Thus, spatial-sampling-induced differences are relatively small compared to that arising from other sources of uncertainty.