Size and shape patterns of fish groups are collective outcomes of interactions among members. Consequently, group-level patterns are often affected when any member responds to changes in their internal state, external state, and environment. To determine how groups of fish respond to components of their physical and ecological environment, and whether the response is influenced by a component of their external state (i.e., fish age), we used a multibeam system to collect three-dimensional grouping characteristics of 5 age categories of juvenile walleye pollock (age 1, age 2, age 3, mixed ages 1 and 2, and mixed ages 2 and 3) across the eastern Bering Sea shelf over two consecutive years (2009–2010). Grouping data were expressed as metrics that described group size (length, height), shape (roundness, spread), internal structure (density, internal heterogeneity), and position (depth, distance above bottom). Physical data (water temperature measurements) were collected with temperature-depth probes, and ecological data (densities of predators and prey − adult walleye pollock and euphausiids, respectively) were collected with an EK60 vertical echosounder. Juvenile pollock maintained a relatively constant shape, size-dependent density (number fish/mean body length3), and internal horizontal heterogeneity among age categories and in the presence of predators and prey. There were changes to group structure in the face of local physical forcing. Groups tended to move towards the seafloor when bottom waters became warmer, and groups became vertically shorter, denser, and had more variation in horizontal internal density as group depth increased. These results are explored in relation to the value and limitations of using multibeam data to describe how external and internal group structure map onto environmental influences.