We are at a unique time in the study of our place in space. On one hand, we operate in the same paradigm that has guided the study of space science for the past couple of decades, and on the other a rising dependence of our economic and social well-being on space demands a shift. Everywhere in our society ‘big data’ (defined by four V’s: volume, variety, veracity, and velocity) and the advent of sophisticated and efficient methods to explore these data (i.e., data science) present new opportunities for discovery, and the time is ripe for these methods to shift how we study the physics of space. We will first discuss the meaning of data science in the context of space science, and then demonstrate the potential for new discovery through a power use case: leveraging Global Navigation Satellite Systems (GNSS) signals for space weather prediction. In this use case, we take advantage of a large volume of data from GNSS signals, data science-driven technologies, and a machine learning algorithm known as the Support Vector Machine (SVM) to develop a novel predictive model for high-latitude ionospheric phase scintillation. This talk will conclude with a perspective on opportunities in space science through ‘big data’ and creating new scientific discovery at the intersection of traditional approaches and data science-driven innovation.