Because of logistical barriers to observing roots, many studies rely on aboveground plant parts, even though their belowground characteristics may be more illuminating. For instance, neglecting roots has resulted in underestimates of global carbon stocks, plant primary production, length of growing season, and plant diversity. Excised roots, such as those found in soil cores, can be best identified using molecular methods. Of these DNA-based methods, fluorescent amplified-fragment length polymorphisms (FAFLPs) are accurate, inexpensive, and able to analyze mixed-species samples without the time-intensive bioinformatics associated with high-throughput sequencing. Here, FAFLP size profiles are determined for boreal forest species. Multiple leaf tissue samples from 214 species common to the boreal forest in northeastern Alberta, Canada were collected. Collection sites were chosen from across northeastern Alberta to represent a range of natural and disturbed habitats. DNA was extracted using modified 2% CTAB protocol and amplified. Fragment lengths of three non-coding cpDNA regions, trnT-trnL intergenetic spacer, trnL intron, and trnL-trnF intergenetic spacer, were resolved using capillary electrophoresis.
Preliminary results show that fragment sizes ranged from 486-1281 bp, 338-700 bp, and 214-535 bp, for the trnT-trnL intergenetic spacer, trnL intron, and trnL-trnF Intergenetic spacer, respectively. Consistent with previous work, the trnT-trnL intergenetic spacer is shown to be the largest and most variable region, but also the most difficult to amplify (amplicon sizes were recovered from only 63% of species). Unique size profiles useful for identification could be created for most species in this study, suggesting that proper identification from excised roots for boreal species is possible using FAFLPs. However, it was not always possible to create unique size profiles for congenerics or closely related species. This problem can be reduced if above ground species surveys are used to limit the reference list of size profiles.