Mixing is an important control on how the oceans function; yet our understanding of ocean mixing is significantly limited by a complex variability in mixing rates and processes and by a paucity of mixing measurements. In the Arctic Ocean, the challenges are substantial: measurements are extremely rare, and latitude, ice and stratification make mixing dynamics unique. In this talk, I’ll discuss ways we are “filling in the map” of Arctic Ocean mixing rates and understanding what sets its space-time geography using data mining and underwater robots. I’ll also show results from experiments with realistic ocean models to argue that this map has important implications for our understanding of ocean functioning and our ability to make robust predictions of future climate change.