Abstract. The energy balance of eddy covariance (EC) flux data is typically not closed. The nature of the gap is usually not known, which hampers using EC data to parameterize and test models. The present study elucidates the nature of the energy gap of EC flux data from winter wheat stands in southwest Germany. During the vegetation periods 2012 and 2013, we continuously measured, in a half-hourly resolution, latent (LE) and sensible (H) heat fluxes using the EC technique. Measured fluxes were adjusted with either the Bowen-ratio (BR), H or LE post-closure method. The adjusted LE fluxes were tested against evapotranspiration data (ETWB) calculated using the soil water balance (WB) method. At sixteen locations within the footprint of an EC station, the soil water storage term was determined by measuring the soil water content down to a soil depth of 1.5 m. In the second year, the volumetric soil water content was also continuously measured in 15 min resolution in 10 cm intervals down to 90 cm depth with sixteen capacitance soil moisture sensors. During the 2012 vegetation period, the H post-closed LE flux data (ETEC = 3.4 ± 0.6 mm day−1) corresponded closest with the result of the WB method (3.3 ± 0.3 mm day−1). ETEC adjusted by the BR (4.1 ± 0.6 mm day−1) or LE (4.9 ± 0.9 mm day−1) post-closure method were higher than the ETWB by 20 and 33%, respectively. In 2013, ETWB was in best agreement with ETEC adjusted with the H post-closure method during the periods with low amount of rain and seepage. During these periods the BR and LE post-closure methods overestimated ET by about 30 and 40%, respectively. During a period with high and frequent rainfalls, ETWB was in-between ETEC adjusted by H and BR post-closure methods. We conclude that, at most vegetation periods on our site, LE is not a~major component of the energy balance gap. Our results indicate that the energy balance gap other energy fluxes and unconsidered or biased energy storage terms.