Tree and shrub roots in sewer pipes

In the urban environment, conflict below the ground exists between tree roots and services – namely, sewer pipes. Because tree roots will grow up a moisture gradient, and pipes both contain moisture within and may collect condensation on the outer surface, tree roots are drawn to the pipes and, in many cases, are able to intrude into the pipe network (through small openings; usually at junctions, and when the pipes are old and made of clay). Such intrusion obviously causes issues, with regards to the blockage of the pipes, and their subsequent fixing. In response to this, the aim of the authors in this study was to assess what urban tree and shrub species’ roots were found within pipes, whether different species and cultivars had different rates at which their roots were able to enter into the pipes, and if the material the pipe was made of impacted upon the rate of root intrusion.

A CCTV image of tree roots blocking a pipe entirely. Source: South Gippsland Water.

Underground sewage pipes were (prior to the study, from 1970-2007) inspected – via CCTV cameras – in the Swedish cities of Malmö and Skövde, and the number of root intrusions along pipe lengths were established and mapped as a result (a total of 2,180 intrusions along 33.7km of pipe). From these records, the authors located survey sites. These surveys also noted what the pipe was made of (concrete or PVC), the pipe’s date of construction, and the pipe dimensions. In relation to the plants featured in the study, a 2008 survey of the tree and shrub populations in both cities led to 4,107 individuals being identified within a 20m radius from the pipes where intrusions had been located. Data from earlier inventories enabled the authors to expand the research to 14,552 trees and shrubs.

Once the tree and shrub locations had been plotted against locations of root intrusions into pipes, the trees were segmented into 186 different genera, species, and cultivars. As this was considered a vast range, the authors sought to narrow-down the list to a more manageable number. This was achieved by selecting only those trees and shrubs within 10m of an intrusion point, where no other vegetation was found within the original 20m distance; though if two individuals of the same species were found within a 20m distance from an intrusion point then the nearest tree of the two would feature as part of the survey.

After the final tree and shrub list (comprising of 2,4,21 individuals from 52 different species and cultivars) was drawn up and the data relating to pipe intrusions analysed, it was found that roots of both broadleaved and coniferous species were able to intrude into pipes. Curiously, it was the PVC pipes (0.661) that had a greater rate of intrusion than concrete ones (0.080) per joint (one length in between two joins). The below table outlines species included within the study, and the rate of root intrusion. Evidently, there are many tree species featured, and this was after many had already been filtered-out of the study scope.

Data captured within the survey for each tree species (or genus) relating to the rate at which roots intruded into pipes.

From the results gathered, the authors remark that the rate at which Malus floribunda entered into pipes was very significant. Compared to previous studies, and even compared to other species of Malus, this species could be considered very able in terms of root intrusion. In fact, so great was its ability to enter into pipes that it trumped willows at a rate of 3:1. At a broader level however, what this study shows is that many tree and shrub species have similar rates of intrusion into sewer pipes, which therefore suggests that discrimination against particular species may not necessarily be wholly justified.

Building on the above comment, we can observe that Tilia cordata and some species of Ulmus also intrude into pipes quite significantly, and even more so than the Salix species observed during the study. Not surprisingly however, Populus canadensis was found to have roots within pipes more often than most other species surveyed (asides from Malus floribunda). Granted, other Populus species were almost half as likely to enter pipes, in light of this survey data. Therefore, it’s not simply a case of observing particular genera that infiltrate regularly into pipes, but particular species within a genus. This seems constant throughout, though Acer spp., Malus spp., Populus spp., and Ulmus spp. most effectively demonstrate this.

Interestingly, the lower than expected rate of root intrusion by Populus spp. and Salix spp., the authors allege, is because previous studies have obtained results with a disproportionately high number of thse two genera across the survey sites. As a result, it may simply have been a case that, because these two genera featured so heavily, their roots were more often found in pipes than other genera and species were. If populations of each tree species were normalised, it may have been found that other tree genera and species intruded into pipes more frequently, potentially. This is, in fact, what this study shows, as it was the mean number of root intrusions per pipe joint that were calculated (instead of just the number of intrusions).

Granted, this study was not without its limitations. The authors even acknowledge this, and state that tree and shrub species present on a site is not the only determining factor to do with intrusion rate. Soil properties, what pipes are made out of, their age, condition, and the distance to the nearest pipe from a tree are also almost certainly to influence upon infiltration rate.

This study is nonetheless serious food for thought, and consideration should of course be given as to what species are to be planted in areas near to pipes. Hopefully, this research will aid with the decision-making process (or, perhaps, the opposite!).

Source: Östberg, J., Martinsson, M., Stål, Ö., & Fransson, A. (2012) Risk of root intrusion by tree and shrub species into sewer pipes in Swedish urban areas. Urban Forestry & Urban Greening. 11 (1). p65-71.

To discuss this post, please leave either a message below or post over on Arbtalk.

Advertisement