In June 2013, a test of sample processing steps was undertaken for limnological/oceanographic sampling of total organic carbon (TOC), particulate organic carbon (POC), dissolved organic carbon (DOC), and colored dissolved organic matter (CDOM). This was to determine the magnitude and sources of errors in the sampling of these organic carbon field measurements. Replicate Niskin rosette casts in Lake Superior (with a difference of 0.79 km in site location and 1 hour 6 min in sampling time) were found to be significantly different at both alpha= 0.05 and 0.10 for DOC, TOC, e2/e3, and CDOM, but not POC. The TOC, DOC and POC average concentration and the standard deviation of all replicates of the two casts was 2.3 +/- 0.2, 2.3 +/- 0.3, and 0.14 +/- 0.02 mg/L respectively. The variation due to sample handling within one cast for DOC and TOC is 0.1 mg/L or 5% relative standard deviation. For POC, the variation is 0.02 mg/L or 14%. Use of different Niskins from the same cast did not cause significant effects.