Salt water intrusion often troubles ecosystems and municipalities. Several fields of engineering and science have quantified and described these flows under varying conditions with both physical and numerical modeling. However, no investigation has been performed into the effects of complex longitudinal channel geometry on a propagating salt water intrusion front. This study was undertaken in order to examine this situation and develop a relationship between longitudinal channel geometry and important parameters of flow. A predominantly transparent open channel flume was built that could modularly form several curved channel geometries. This flume was attached to a reservoir fitted with a release gate. The reservoir was filled with salt water and the flume with fresh water, both at equal heads. The gate was released and the intruding wedge was tracked via video recording. The data was then correlated using dimensionless parameters. The apparatus was verified using a straight channel setup and comparing it to data gathered by previous researchers. Additionally, a full experimental uncertainty analysis was performed to show the accuracy with which the data was taken Fifty-six experiments were performed. For each channel geometry four experiments were performed in duplicate under varying conditions (10-ppt, 25-ppt, 60-ppt, and 25-ppt with rough channel bottom). There were 7 experimental setups including large (22.5-inch centerline) radius 180-degree bend, small radius (12.5-inch centerline) radius 180-degree bend, 90-degree large and small bend, S-curve large and small bend, and straight channel. The results in the straight channel experiments not only verified the setup versus previous researchers, but also served as controls for the curved channel geometry experiments. Comparing the results of the straight channel experiments to each of the curved channel geometries showed that there were no measurable differences within the experimental uncertainty
