Isolates of the predominantly asexual Phytophthora infestans, the oomycete causal agent of potato late blight, population in the US are characterized by clonal lineage, or asexual descendants of a single genotype. Current tools for clonal lineage identification are time consuming and require laboratory equipment. We previously found that foliar spectroscopy can be used for high accuracy pre- and post-symptomatic detection of P. infestans infections caused by clonal lineages US-08 and US-23. In this work, we find that there are subtle but distinct differences in spectral responses of potato foliage infected by these clonal lineages in both controlled, time course (12 to 24-h intervals over 5 days) growth chambers experiments and naturally-infected samples from commercial production fields. In both settings we measured continuous visible to shortwave infrared reflectance (400-2500 nm) on leaves using a portable spectrometer with contact probe. We consistently discriminated between infections caused by the two clonal lineages across all stages of disease progression using partial least squares discriminant analysis (PLS-DA) with total accuracies ranging from 88 to 98%. Three-class random forest differentiation between control, US-08, and US-23 yielded total discrimination accuracy ranging from 68 to 76%. Differences were greatest during pre-symptomatic infection stages and progressed towards uniformity as symptoms advanced. Using PLS-regression trait models we found that total phenolics, sugar, and leaf mass per area were different between lineages. Shortwave infrared wavelengths (>1100 nm) were important for clonal lineage differentiation. This work provides a foundation for future use of hyperspectral sensing as a non-destructive tool for pathovar differentiation.