Many complex ecological and social-ecological systems are capable of nonlinear feedbacks. These can result in abrupt and unanticipated shifts in the dynamical regime of a system, as environmental conditions move the system beyond a tipping point. Research in early warning signals of tipping points focusses on ways to predict these tipping points ahead of time by looking for telltale signatures of noise in the data before the tipping point is reached. In this talk, I will describe some research research in our group that (1) characterizes how conventional early warning signals in ecological systems change in the face of social-ecological feedbacks, and (2) explores new types of early warning signals that predict not only the presence, but also the type, of tipping point that is being approached. I will also discuss some opportunities to find early warning signals of social-ecological transitions in social media data, such as tweets on climate change and vaccines.