Performance-Based Hydraulic Conductivity Models for Unbound Granular Materials

Specifications for unbound granular materials (UGM) must be based on the function of the
unbound layer (drainable layer, high stiffness, or both). Using performance-based specifications
that are based on laboratory performance of UGM (resilient modulus, permanent deformation,
and permeability) provides durable and longer lasting unbound layers. This paper investigates
the effect of fines content variation on the permeability (hydraulic conductivity) of two types of
UGM, gravel and 100% crushed limestone. Hydraulic conductivity tests were conducted on
compacted gravel UGM with 4.0% fines, 9.0% fines, and 14.5% fines. For 100% crushed
limestone, hydraulic conductivity tests were conducted on compacted samples with 4.5% fines,
10.5% fines, and 16.0% fines. For gravel, the hydraulic conductivity decreased by 84% due to
increasing fines content from 4.0% to 14.5%. For Limestone, the hydraulic conductivity
decreased by 81% due to increasing fines content from 4.5% to 16.0%. For gradations with
comparable fines content, limestone material showed higher hydraulic conductivity than that for
gravel. The laboratory measured hydraulic conductivity was compared to Level 2 design inputs
in the Mechanistic-Empirical Pavement Design Guide (Pavement ME). The calculated hydraulic
conductivity using Pavement ME Level 2 equation was found to be higher than the laboratory
measured values by 200% to 700% for gravels, and by 25% to 300% for limestones. The
difference between the hydraulic conductivity computed using Pavement ME Level 2 equation
and the laboratory measured values increased as the fines content increased. Overestimating the
hydraulic conductivity of UGM layers may result in under designed pavement structures and
consequently reduced pavement service life. The research recommends calibrating the
performance based models with locally generated hydraulic conductivity data to improve the
reliability of the models.