These three conferences are influential in my academic GIScience circles.
(1) Conference on Spatial Information Theory; http://www.cosit.info/
(2) International Conference on Geographic Information Science (GIScience); http://www.giscience.org/
(rising in importance)
(3) ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems; http://www.sigspatial.org/sigspatial-conferences
An accepted paper for COSIT or GIScience might be somewhat equivalent to publication in International Journal of Geographic Information Science or Transactions in GIS (IJGIS Impact Factor in 2012: 1.613 & TGIS Impact Factor in 2012: 0.906).
Of course, the interdisciplinary spatial journals have the highest impact:
Due to the specialized nature of GIScience, I expect the most-influential papers on geospatial information research will begin to be published in journals with a wider reach. That same gravity also will pull the work to higher impact conferences, such as one of the Association of Computing Machinery specialty meetings -- or perhaps to large conferences, like the American Geophysical Union Fall Meeting.
Simultaneously, the Internet minimizes the need for singular important conferences and journals. In the longer term, I expect closed, expensive, or proprietary channels to be diminished in importance.