DNA analysis of soil extracts can be used to investigate fine root depth distribution of trees

Knowledge of a tree species or cultivar's fine root distribution is important. However, the time and resource requirements of established soil core based methods where live from dead root determination is required, act as a constraint to large studies. We developed a method to determine live fine root DNA density for mango (Mangifera indica). Soil-root samples had large roots separated by sieving (min. 2 mm aperture) and soil DNA extractions made on the sieved soil containing fine root fragments that had passed through the sieve. We showed that the DNA yields of these samples could determine fine root distribution. Abstract Understanding the root distribution of trees by soil coring is time-consuming as it requires the separation of roots from soil and classification of roots into particular size classes. This labour-intensive process can limit sample throughput and therefore sampling intensity. We investigated the use of quantitative polymerase chain reaction (qPCR) on soil DNA extractions to determine live fine root DNA density (RDD, mg DNA m−2) for mango (Mangifera indica) trees. The specificity of the qPCR was tested against DNA extracted from 10 mango cultivars and 14 weed species. All mango cultivars and no weeds were detected. Mango DNA was successfully quantified from control soil spiked with mango roots and weed species. The DNA yield of mango root sections stored in moist soil at 23–28 °C declined after 15 days to low concentrations as roots decayed, indicating that dead root materials in moist soil would not cause false-positive results. To separate large roots from samples, a root separation method for field samples was used to target the root fragments remaining in sieved (minimum 2 mm aperture) soil for RDD comparisons. Using this method we compared the seasonal RDD values of fine roots for five mango rootstock cultivars in a field trial. The mean cultivar DNA yields by depth from root fragments in the sieved soil samples had the strongest relationship (adjusted multiple R2 = 0.9307, P < 0.001) with the dry matter (g m−2) of fine (diameter