By Ian Longley
Outside our major centres, New Zealand is a country of many small towns separated by long drives. The harder somewhere is to get to, the more likely it is tucked away in a valley or basin, and the more likely it is to rely on wood for home heating. How many have air quality that meets the current NESAQ, let alone any future PM2.5 standard?
The truth is, for many towns, we don’t know.
The problem is the need to monitor a lot of far-flung locations with only a small population and hence small budgets based on limited rates. It is no surprise that many towns never get monitored and their air quality status remains a mystery.
And even if monitoring can be funded, where should it be placed? Concentration gradients in small towns can be stark, varying over very short distances. Choosing the “wrong” site (and maybe only a single site is available) that severely under-represents typical concentrations can almost render the whole exercise pointless.
Low-cost sensors offer to solve both of these problems simultaneously. Not only is the cost of a monitor one to two orders of magnitude less than a conventional regulatory monitor, but the demands on the site are much reduced. There is no power cost, and they are quicker to deploy (reducing labour cost). They are also easily deployed on almost any street.
We’ve been working hard over the years to streamline monitoring with low-cost sensors to ensure that the capital cost advantage is not undermined by higher operational costs. We’ve also been keen to find ways that low-cost sensors can be used to provide more information than a conventional approach for less cost and make the whole process as simple and reliable as possible.
One of our most recent examples was our work with Horizons Regional Council.
The Manawatu-Whanganui region is a classic case of a large number of scattered small towns, most of which have no history of air monitoring. Council staff were considering purchasing an instrumented trailer to slowly “tour” some of these towns, an approach we’ve come across in several regions. We suggested an alternative approach.
Horizons funded a pilot in the small town of Woodville (population )1,680). So last winter we provided Horizons with 10 “Clarity” air monitors and our suggestion of the most effective locations to place them. Council staff did the deployments (on street light poles). We kept an eye on monitor performance. At the end of the 3.5-month campaign we were delighted to report a 100 % data recovery rate – a strong indication of how far low-cost sensing has come in terms of reliability.
By using a grid of sensors instead of a single monitor we were able to observe strong spatial variation with the most polluted area being 1.5 to 2 times more polluted than the least on nearly every day of our observational campaign, and that the spatial pattern was quite consistent from day to day and across the day. We were also able to show that combustion sources dominated with minimal impact from dusts.
In terms of whether Woodville meets the current NESAQ, or alternative limits or targets, this approach does not provide a definitive answer. However, we delivered an indicative answer in two ways. Firstly, we adjusted the Clarity data using equations derived from co-location experiments that we have conducted in a number of locations. This still contains some uncertainty. Secondly, we compared the data from Woodville with data we previously collected with a Clarity monitor at the Otago Regional Council site in Alexandra. The range of concentrations for comparable times of year were similar between the two towns, implying Woodville’s air quality might be similar to that reported from the Alexandra site. Both of these pieces of evidence indicated a high risk that air quality in Woodville exceeds the NESAQ.
From our perspective as the service provider, a key area of investment has been on the automation of much of the QA process and data analysis. This also substantially reduces the cost of the whole exercise whilst also increasing the value that we can derive from the data collected.
Our recommendation to Horizons was to follow up the first study with a second in which a full 12 months’ worth of data is collected in Woodville to more formally establish the state of air quality. This can now be conducted at a single site, but, because of the first stage study, Horizons now have a very robust understanding of where that single site needs to be to represent the airshed needs to be located, or the adjustment required if an alternative location is used.
This whole process cost less, required less technician time (2 working days) and was quicker to execute than the original trailer-based proposal. Furthermore, it yielded more information and produced a much more robust and less uncertain understanding of how the Woodville airshed works.
