The 2015/2016 El Niño event caused severe changes in precipitation across the tropics. This impacted surface hydrology, such as river run-off and soil moisture availability, thereby triggering reductions in gross primary production (GPP). Many biosphere models lack the detailed hydrological component required to accurately quantify anomalies in surface hydrology and GPP during droughts in tropical regions. Here, we take the novel approach of coupling the biosphere model SiBCASA with the advanced hydrological model PCR-GLOBWB to attempt such a quantification across the Amazon basin during the drought in 2015/2016. We calculate 30-40% reduced river discharge in the Amazon starting in October 2015, lagging behind the precipitation anomaly by approximately one month and in good agreement with river gauge observations. Soil moisture shows distinctly asymmetrical spatial anomalies with large reductions across the north-eastern part of the basin, which persisted into the following dry season. This added to drought stress in vegetation, already present owing to vapour pressure deficits at the leaf, resulting in a loss of GPP of 0.95 (0.69 to 1.20) PgC between October 2015 and March 2016 compared with the 2007-2014 average. Only 11% (10-12%) of the reduction in GPP was found in the (wetter) north-western part of the basin, whereas the north-eastern and southern regions were affected more strongly, with 56% (54-56%) and 33% (31-33%) of the total, respectively. Uncertainty on this anomaly mostly reflects the unknown rooting depths of vegetation.This article is part of a discussion meeting issue 'The impact of the 2015/2016 El Niño on the terrestrial tropical carbon cycle: patterns, mechanisms and implications'.