Estimating lake evaporation is a challenge due to both practical considerations and theoretical assumptions embedded in indirect methods. For the first time, we evaluated measurements from an optical microwave scintillometer (OMS) system over an open water body under arid conditions. The OMS is a line-of-sight remote sensing technique that can be used to measure the sensible- and latent heat fluxes over horizontal areas with path lengths ranging from 0.5-10 km. We installed an OMS at a saline lake surrounded by a wet-salt crust in the Salar del Huasco, a heterogeneous desert landscape in the Atacama Desert. As a reference, we used Eddy Covariance systems installed over the two main surfaces in the OMS footprint. We performed a footprint analysis to reconstruct the surface contribution to the OMS measured fluxes (80% water and 20% wet-salt). Furthermore, we investigated the applicability of the Monin-Obukhov Similarity Theory (MOST) which was needed to infer fluxes from the OMS-derived structure parameters to the fluxes. The OMS structure parameters and MOST were compromised, which We mitigated by fitting MOST coefficients to the site conditions. We argue that the MOST deviation from values found in the literature due to the effects of the surface heterogeneity and the non-local processes induced by regional circulation. With the available data set we were not able to rule out instrument issues, such as additional fluctuations to the scintillation signal due to absorption or the effect of vibration in high wind conditions. The adjusted MOST coefficients lowered by a factor of 1.64 compared to using standard MOST coefficients. For H and K=LvE, we obtained zero-intercept linear regressions with correlations, R2, of 0.92 and 0.96, respectively. We conclude that advances in MOST are needed to successfully apply the OMS method in landscapes characterized by complex heterogeneity such as the Salar del Huasco.