Proposed formulations for error reduction in leachate pollution index (LPI) estimation due to the absence of leachate parameters

Abstract

The applications of leachate pollution index (LPI) as an environmental index to express the overall leachate-contaminating ability of landfills are increasing. Majority of these applications wrongly quantify LPI due to the absence of various leachate parameters. The traditional linear weighted aggregation equation used in LPI estimations, result in larger errors as missing parameters increase. In this article, error reduction equations were established to predict LPI values more accurately. Approximately 1797 unique combinations of leachate weights ( ∑ wi) were randomly assigned to cover the missing parameters and augment the accuracy of the proposed error reduction equations. The upper and lower boundaries of LPI were accurately estimated in each case, starting from 1 missing parameter to 12 missing parameters. Various linear equations were generated, when possible, to ease the calculations. Subsequently, the arithmetic, geometric, and harmonic averages were applied between the LPI limits in five different case studies to demonstrate the validity of the developed error reduction equations. These equations were capable of providing reliable LPI values even when up to twelve leachate pollutants were absent. It was found that the harmonic average was the best average to be applied between the LPI boundaries, which mostly resulted in errors around ± 20% for ∑ wi values up to 0.3. • Modified LPI formulations for error reduction of leachate pollutants absence. • Random data generation was applied to cover most possible LPI combinations. • Harmonic average outperformed peer arithmetic and geometric averages. • Reliable values of LPI can be estimated mostly with errors up to ± 20%.