Winter limits resource availability in temperate lakes, exerting a strong influence on the northern distributions of many temperate species by altering the recruitment of age-0 fishes. Recently, there has been growing concern that projected climate change could alter over-winter survival of many fish species and have ecosystem-level effects on all trophic levels of aquatic ecosystems. Oneida Lake has been the focus of a long-term data set across several trophic levels and provides the ideal location to study over-winter mortality in fishes. In Chapter 1, I provide a review of current methods used throughout the literature to address questions involving over-winter mortality in fishes. Specifically, I address methods used, their pros and cons, and what has been learned by the application of each method. Common methods evaluated include experiments, field observations, experiments and field observations, and analysis of long-term data sets. Given that mortality is commonly driven by complex interactions between multiple factors, I suggest researchers use multiple approaches to study fish over-winter mortality. In Chapter 2, I conducted a series of experiments and field sampling to develop a conceptual model of gizzard shad over-winter mortality in Oneida Lake. Gizzard shad exhibited high mortality rates as water temperatures declined prior to ice formation and are highest at temperatures less than 4 degrees C. Habitat sampling demonstrated that shad congregate in high densities in Oneida Lake marinas, which provide a nearshore temperature refuge but may become anoxic during years with consistent ice cover. Within any given year, shad survival is likely a function of length entering winter, rate of temperature decline preceding ice-on, and ice duration.