Knowledge on groundwater-surface water interaction and especially on exchange fluxes between streams and aquifers is an important prerequisite for the study of transport and fate of contaminants and nutrients in the hyporheic zone. One possibility to quantify groundwater-surface water exchange fluxes is by using heat as an environmlental tracer. Modern field equipment including multilevel temperature sticks and the novel open-source analysis tool LPML make this technique ever more attractive. The recently developed LPML method solves the one-dimensional fluid flow and heat transport equation by combining a local polynomial method with a maximum likelihood estimator. In this study, we apply the LPML method on field data to quantify the spatial and temporal variability of vertical fluxes and their uncertainties from temperature-time series measured in a Belgian lowland stream. Over several months, temperature data were collected with multilevel temperature sticks at the streambed top and at six depths for a small stream section. Long-term estimates show a range from gaining fluxes of -291 mmday -1 to loosing fluxes of 12mmday -1 ; average seasonal fluxes ranged from -138mmday -1 in winter to -16 mmday -1 in summer. With our analyses, we could determine a high spatial and temporal variability of vertical exchange fluxes for the investigated stream section. Such spatial and temporal variability should be taken into account in biogeochemical cycling of carbon, nutrients and metals and in fate analysis of contaminant plumes. In general, the stream section was gaining during most of the observation period. Two short-term high stream stage events, seemingly caused by blockage of the stream outlet, led to a change in flow direction from gaining to losing conditions. We also found more discharge occurring at the outer stream bank than at the inner one indicating a local flow-through system. With the conducted analyses, we were able to advance our understanding of the regional groundwater flow system.