The Colorado River is an important water source in the southwestern United States and northern Mexico. High concentrations of dissolved solids in the river, sourced mainly from the Upper Colorado River Basin (UCOL), cause hundreds of millions of dollars in damages annually to crops and infrastructure. Determinations of total dissolved solids (TDS) in river and tributary samples often rely on summed concentrations of constituents in solution reported as the sum of constituents (SOC), which includes the bicarbonate concentration converted to the equivalent carbonate mass that would be present as a residue after drying. Alternatively, salinity, similar to SOC but including the entire concentration of bicarbonate in solution, may be used as a measure of dissolved solids. Use of SOC results may under-represent actual dissolved-solids concentrations and loads in streams where bicarbonate is a substantial component of the dissolved solids in solution. The objective of this manuscript is to evaluate the differences between the SOC and salinity determinations of dissolved solids in UCOL streams and rivers. Water-quality data from the U.S. Geological Survey were used to compute salinity concentrations at UCOL stream sites for comparison with SOC determinations. Results from 8,001 samples at 418 UCOL sites indicate a median increase in dissolved solids of 20% (13% and 30%, 25th and 75th percentiles, respectively) using the salinity method compared with SOC results. Differences in dissolved solids attributable to the computational approach for handling bicarbonate at UCOL sites were significantly greater than laboratory variability based on results from 890 replicate analyses. Salinity may be a more useful indicator of water quality than SOC in systems with substantial proportions of bicarbonate in the composition of dissolved solids, including the Colorado River and UCOL sites.