Some Observations Based on Complementary International Evaluations of Edar Vehicle Emissions Remote Sensing Technology

Here we report findings from two complementary blind evaluations of the vehicle emissions measurement capabilities of the EDAR remote sensing system. The first study, by Colorado Department of Public Health and Environment and Eastern Research Group, was a simulated exhaust gas test that used conventional remote system auditing methods to investigate the accuracy of the EDAR. The EDAR measured CO, NO, CH4 and C3H8 concentrations with high linearity, low bias, and low drift over a wide range of concentrations and vehicle speeds. Instrument accuracy was high (R2 0.996 for CO, 0.998 for NO; 0.983 for CH4; and, 0.952 or better for C3H8) and detection limits are low (50-100 ppm for CO; 10-30 ppm for NO; 15-35 ppmC for CH4; and, 100-400 ppmC3 for C3H8). The second study, by the Universities of Birmingham and Leeds and King’s College London, used the comparison of EDAR, PEMS and car chaser system measurements collected under real-world conditions to provide a measure of in situ EDAR performance. Given the analytical challenges associated with aligning these very different measurements, the observed degrees of agreement (e.g. EDAR versus PEMS R2 0.924 for CO/CO2; 0.969 for NO/CO2; 0.826 for NO2/CO2; and early observations on PM measurement and car chaser experiments) were all highly encouraging and demonstrate that EDAR also provides a representative measure of vehicle emissions under real-world conditions.