Mixing of a dye tracer in the Delaware plume: Comparison of observations and simulations

In April 2003 we injected Rhodamine‐WT fluorescent dye into the halocline at the base of the Delaware River plume during an upwelling event. The dye immediately bifurcated: One third of the dye moved offshore with a portion of the plume, while the other 2/3 remained within the halocline, near the coastline. This behavior has not been previously observed, and it presents a strong challenge to contemporary coastal ocean models. Here we employ a numerical model of Delaware Bay and the adjacent coastal region that incorporates the Mellor‐Yamada Level 2.5 turbulence closure scheme to simulate the observed dye injection event. Comparison between the simulation and observations reveals that the model is able to successfully reproduce the flow field, simulating the separation and subsequent advection of the dye. Consequently, the simulation is used to examine the relative effects of wind‐ and buoyancy‐driven transport on mixing and advection of the dye. Comparison of simulated and observed plumes reveals that the model accurately reproduces the observed mixing and accompanying vertical salt flux within the offshore plume. Examination of the simulation reveals a unique combination of wind‐ and buoyancy‐driven flow that results in the dye remaining in the halocline of the onshore plume despite significant vertical mixing due to winds. The strong stratification within the inshore plume causes the closure scheme to shut off the vertical diffusivities, leaving the circulation model with arbitrarily chosen constant background values. Varying these values reveals only a weak dependence of the salt flux within the highly stratified regions due to the highly localized nature of the salt flux.