There is pressure from some quarters for new pipelines to be kept out of farmland and laid along roads, as I have mentioned previously. A little while ago I was asked whether the APIA/POG pipeline incident database could reveal anything about the relative safety of pipelines in roads and elsewhere. The outcome was striking.
Consider first all external interference incidents (including both hits and near misses). About 1/3 of all incidents occur within road reserves. If we restrict the data to actual damage incidents then fully half occur within road reserves. I need to qualify this by saying that older data (from before 2000) did not record land tenure so was omitted, which reduces the data available but also means that the data used reflects recent years.
We can take this a bit further by estimating what proportion of pipeline is laid in road reserves as opposed to general cross-country installation. Unfortunately at this stage I can only guess, but it seems reasonable to guess a maximum of 1% of pipeline in roads (and in fact probably a lot less). But if we work with 1%, that suggests that the 1% of pipeline in roads accounts for half of damage incidents. Put another way, the damage rate in roads is 100 times higher than elsewhere if my 1% guess is correct, even higher if the road fraction is smaller.
All that is a really compelling argument to resist locating pipelines within road reserves.