“…Both baseline stability and reproducibility were excellent with typical RSDs (n ס 3) of Ͻ2%. It is possible to thermostat the instrumentation [8] to provide a stable response, but for this field trial a 5 mg L 1מ NO 3 -N standard was used as a reference for all daytime samples. With few exceptions, reproducibility between replicate injections was Ͻ2% RSD (n ס 3), pooled (all three calibrations) RSDs were Ͻ4% (n ס 9), and linear correlation coefficients for all calibrations were good (pooled result: r 2 ס 0.9998, n ס 6).…”