Part 1: A theoretical approachChaotic dynamics have been directly detected in ocean dwelling species, and asymmetrical advective dispersal is the dominant dispersal mechanism in the ocean. Therefore, we must better our understanding of chaotic population dynamics under the influence of a unidirectional current. Here, we examine a spatially explicit, density-dependent population in a unidirectional current, where offspring are dispersed some distance downstream of the parent generation. This system displays a rich variety of dynamics from chaotic to steady-state depending on the distance the offspring are moved downstream, the diffusive spread of the offspring (which can be interpreted as a stochastic component of the advection), and the domain size. We will illustrate how advection can serve as a stabilizing or destabilizing force, depending on its size relative to the domain sizediffusive spread. Improving our understanding of the dynamics of these systems will aide in conservation and sustainable management efforts. Part 2: Nonlinear dynamics in fisheries data versus nonlinear dynamics in fisheries modelsHere, we present an example of directly detected nonlinear dynamics in time series of fish abundance, and illustrate how nonlinearity can be used as a characteristic to assess whether parametric models of complex marine ecosystems are capturing the underlying dynamics of the systems theyre modeling. All models are simplified approximations of reality, but a models utility is contingent on its suitability for a given task. This is particularly true of fisheries models, which are used to inform management of commercially important species. Using nonlinear nonparametric forecasting methods, we will show that raw fisheries data sets have a higher prevalence of nonlinearity and lower predictability than fisheries model outputs, suggesting that model outputs may underestimate variability and overestimate stability. Thus, caution is warranted: using such models for management or scenario exploration may produce unforeseen consequences, especially in the context of unknown future impacts.