In 2008 the Great Lakes Bioenergy Research Center initiated a cropping systems trial at Michigan State University's W.K. Kellogg Biological Station (KBS) and the University of Wisconsin's Agricultural Research Station in Arlington, WI (ARL).  Each site's layout is identical with respect to crops grown and management strategies. Preliminary results demonstrate the importance of site differences on total biomass production, with higher total productivity on the highly fertile prairie-derived silt loams at Arlington, and lower total productivity on forest-derived comparatively less fertile sandy loams at KBS. In spite of the higher productivity at Arlington, perennial biomass crops are still yielding far below published results (5.4 Mg ha^-1 for Switchgrass). This may be because these systems are still in their establishment phase and have not yet reached peak productivity. Climate variability differences between KBS and ARL have also demonstrated potential limitations to the establishment of bioenergy crops in the Upper Midwest. Miscanthus x giganteus rhizomes were effectively established in the summer of 2008 but suffered a 95% stand loss at ARL over the first winter while KBS had no significant loss.  The loss at ARL was likely related to soil temperatures below -3.4°C at rhizome depth during winter. Differences between ARL and KBS are greater among annual cropping systems (13.2 vs. 9.5 Mg ha^-1 for corn grain) than among perennial cropping systems (4.7 vs. 3.6 Mg ha^-1 [excludes Miscanthus]). These differences will likely affect farmer acceptance of dedicated biomass crops on land that is highly productive for cash grain commodities like corn and soybeans. Data from the first three growing seasons provide important insights into the establishment of biomass cropping systems, which should be taken into consideration when evaluating the economic and environmental sustainability of potential biomass feedstocks.