Understanding the many processes controlling peatland ecosystems is beyond the expertise of most, even the exceptionally broad-based Paul Glaser.  Keenly aware of peatland complexities, Paul sought colleagues who could address questions beyond his expertise.  He also did his homework to understand and speak their language and often saw connections that they themselves did not.  Such was the situation when vertical hydraulic gradients and high-resolution GPS, two unrelated technologies, were combined to understand the importance of deep-peat organic decomposition on the global carbon cycle.  Two sites, one each in a raised bog and a fen, were instrumented with pressure transducers installed at multiple depths in columns of peat 4.5 and 3 m thick, respectively.  Vertical hydraulic gradients responded to precipitation, evapotranspiration, changing atmospheric pressure, and changing pressure associated with gas generated by organic decomposition.  GPS stations attached to small trees rooted in the peat measured movement of the peat surface in response to these same processes.  When data from the mid-depth pressure transducer at the bog site started behaving erratically, with pressure increasing suddenly, then stabilizing, then rapidly increasing again, the transducer was obviously beginning to fail; the data were dismissed.  But Paul questioned our dismissal and asked for more robust analysis.  Downloading and analyzing data from both systems led to an aha! moment.  Pressure increased because gas was accumulating beneath an elastic, more competent layer of peat, causing the peat above that layer, as well as the GPS station, to rise.   At some point, typically during low atmospheric pressure, the peat layer ruptured, releasing a large volume of gas, and the peat surface collapsed following ebullition.  During these events, GPS data indicated vertical oscillations of more than 15 cm over a period of minutes to hours.  Gas release based on the pressure-transducer data was calculated to range from 60 to 190 l/m² per event.  The GPS data indicated values ranging from 90 to 120 l/m².  These values were about 10 times larger than data from in-situ flux chambers and indicated that ebullition from peatlands was a globally substantial source of natural carbon to the atmosphere that had not previously been measured.