A dispute between Chile and Bolivia regarding the status and use of the waters of the Silala River, resulted in proceedings before the International Court of Justice, initiated in 2016. The magnitude of the effect of historical channelization in Bolivia on surface water flows emerged as a major point of disagreement. Based on modeling by the Danish Hydraulic Institute (DHI), using the MIKE‐SHE and MIKE‐11 modeling systems, Bolivia suggested initially that the channels had increased surface flow by 30%–40% and later by 11%–33%. In the opinion of Chile's international experts, these effects would be small. This paper reviews the use of DHI's models by Bolivia and, subsequently, by Chile. Concerns about Bolivia's modeling raised by Chile included the selection of boundary conditions, inconsistent use of the two models for different scenarios, the use of inconsistent topographies for different scenarios, unexplained additions of water and numerical instabilities affecting the results. To investigate these discrepancies, additional simulations were performed by Chile. DHI's MIKE‐SHE and MIKE‐11 models were used for all scenarios, with consistent topography, spring and surface water flow representation. Numerical instabilities were reduced by adjustment of time steps and channel topography. These simulations showed that removing the channels reduces surface flow by a maximum of 3.5 L/s, or 2.4%. The sum of the groundwater and surface water outflow decreases by 2.6 L/s or 1.0%. This small decrease lies within the modeling error but is consistent with the expected effect of minor changes in evaporation.This article is categorized under: Science of Water > Hydrological Processes Human Water > Rights to Water Science of Water > Methods

No abstract