What is the spatial scale of monitoring conducted at an individual CCAMLR Ecosystem Monitoring Program (CEMP) site? Answering this question is key to understanding how CEMP data could be used in a feedback management strategy and for identifying critical gaps in monitoring effort where new monitoring effort may be useful. Toward this goal, we investigated data sets from two Pygoscelid penguin species that are monitored at three sites on King George Island/Isla 25 de Mayo, within 30km of one another. We used five indices that fall under three main categories of census (breeders and chicks), reproductive success (crèche rates), and chick growth (fledge weights). We found strong positive correlations across sites in census data, implying similar information is being collected at all three sites. We also found evidence of and site- and species-specific differences that highlight heterogeneity in indices of reproductive success and chick growth on local scales. Heterogeneity on such a small spatial scales suggests the need for CEMP monitoring to be distributed widely to encapsulate population responses to changing environments and fishing activity. Within a broad network of CEMP monitoring, it may also be useful to have several monitoring clusters like that on King George Island to help identify the relative importance of local environmental factors and better estimate the range of variability that such factors can introduce into CEMP indices; partitioning such variability seems critical for resolving environmental and fishery impacts on monitored predators.