Our manuscript analyses the surface runoff variability, and its controlling factors in a small basin with gentle slopes, at the headwaters of a flat catchment, to improve the knowledge of the hydrology of plain areas under agriculture. We study runoff, rainfall and antecedent conditions in the argentine Pampas region. We use correlations, regressions and quantitative and qualitative descriptive information of the system: erosion signs, ground cover by crops, groundwater depth data and temporal changes in the drainage network, to discuss and understand the complexity of the runoff process by frameworks to study (dis)connectivity. The analysis of 56 events evidenced a nonlinear rainfall–runoff relationship. In contrast with other works, we identified clear upper limit events, under which hydrological responses emerge, as a result of combinations of antecedent wetness, rainfall erosivity, ground cover and preferential drainage paths. We separated the nonlinear rainfall–runoff response in three linear relationships according to differences in antecedent wetness conditions. We found differences in runoff responses under wet and dry antecedent conditions, but complex responses under medium antecedent conditions. The analyses of the inputs, the structural and the functional elements of the (dis)connectivity frameworks, were key in the understanding of the temporal changes of runoff, and its complex responses. Temporal coincidences of connectivity components and their feedbacks appear to be strongly associated with the runoff dynamics. High‐magnitude hydrological responses occur with complete coincidences, while partial coincidences between the components reduce connectivity and low magnitude and/or heterogeneous responses prevail. Thus, these analyses suggest that runoff is controlled by (dis)connectivity in this basin with gentle slopes. Our work contributes to the understanding of the process of surface runoff in the context of humid flatlands under agricultural land use, by the identification of the complex combinations of factors which regulate/control the (dis)connectivity that helps to interpret the nonlinearities of runoff.