Project Canary (PC) presents the results of its third series of field testing of methane and VOC emissions from Colorado State University’s Methane Emissions Technology Evaluation Center (METEC). The main purpose of this round of testing is to continue improving our quantification analytics platform to localize emitting sources accurately and calculate their emission rates. We deployed eight Canary X sensors and Project Canary’s analytics platform to localize and quantify these unknown leak rates. Results show a cumulative flux error of -4.99% over 45 experimental methane releases. In the third round of METEC testing, our quantification models were improved upon from their previous iterations. We have employed new methods for filtering, background calculation of methane, and data post-processing, among other new features. For example, the grouping of experiments mitigates the effect of outliers by providing a normal-like distribution of fluxes. A normally distributed set of fluxes improves averaging since there will be more samples, and the outliers will receive less weight. Our results demonstrate that the addition of these statistical methods improved the accuracy of the quantification algorithm.