9.1 Time formats
In September (last Sunday of the month) each year, New Zealand switches to New Zealand Daylight Time, moving clocks forward one hour. This reverts to New Zealand Standard Time (NZST) in April (first Sunday of the month), when clocks move one hour back (Department of Internal Affairs, 2009).
For all air quality data, it is recommended that time and date always be reported in NZST. Converting to daylight time would upset the averaging processes. It is also recommended that the midnight hour be labelled as 24.00 instead of 00.00 to avoid time averaging issues.
Recommendation 27: Time format
Times and dates should always be reported in New Zealand Standard Time.
Labelling of the midnight hour should be 24.00 instead of 00.00.
9.2 Reporting data formats
Instruments have different precision levels and those undertaking air quality monitoring use different instruments. This leads to inconsistencies in reporting data at the national level. To address this, it is recommended that a uniform format be used for reporting air quality data.
Recommendation 28: Reporting data formats
It is recommended that data from all monitoring sites be reported in the following format:
|Contaminant||Data format||Nominal precision|
|Carbon monoxide||X.X mg/m3||tenth of a mg/m3|
|Nitrogen dioxide||X.X µg/m3||tenth of a µg/m3|
|Ozone||X.X µg/m3||tenth of a µg/m3|
|Particulate matter (PM10)||XX µg/m3||a whole µg/m3|
|Sulphur dioxide||X.X µg/m3||tenth of a µg/m3|
|Lead||X.X µg/m3||tenth of a µg/m3|
9.3 Significant digits and rounding protocols
Significant digits are specified in the recommended data format in section 9.2. As an example, PM10 should be reported as a whole number, while NO2 should be reported up to the first decimal place. If instrument precision is higher than the recommended data format, it would be necessary to round off to the significant digit when reporting results.
When the value following the significant digit is equal to or greater than 5, the digit should be rounded up; otherwise, the digit is retained (National Environment Protection Council, 2001).
49.9 is rounded-off to 50 µg/m3
50.3 is rounded-off to 50 µg/m3
50.5 is rounded-off to 51 µg/m3
50.7 is rounded-off to 51 µg/m3
199.99 is rounded-off to 200.0 µg/m3
200.04 is rounded-off to 200.0 µg/m3
200.05 is rounded-off to 200.1 µg/m3
200.08 is rounded-off to 200.1 µg/m3
The recommended data format in section 9.2, and the rounding-off protocols above also provide guidance for the reporting of all exceedences. For all monitoring sites, an exceedence occurs when the reported concentration is above the standard, after rounding to the significant digit (see Recommendation 28). For example, a PM10 exceedence is counted once a concentration of 50.5 µg/m3 is obtained.
Recommendation 29: Significant digits and rounding-off protocols
The recommended data format in section 9.2 specifies the significant digits for the reporting of a contaminant.
When the value following the significant digit is equal to or greater than five, the digit should be rounded up; otherwise, the digit is retained.
For all monitoring sites, an exceedence occurs when the reported concentration is above the standard, after rounding to the significant digit.
9.4 Summary statistics
Air quality monitoring can generate substantial amounts of data. Different audiences will be interested in different levels of detail. Most users, however, like to get high-level information that describes air quality conditions during a particular period. This is why summary statistics are useful in reporting air quality data.
It is recommended that the following summary statistics be prepared for each reporting period:
maximum and minimum concentrations based on the relevant averaging period
mean (arithmetic) and median
number of exceedences
percentiles (eg, 10th, 25th, 50th, 75th, 99.7th)
proportion of results belonging to particular bands (eg, air quality categories).
Table 4 is an example showing summary statistics of 24-hour average concentrations for PM10 in a year. Figure 9.1 shows an example of a graphical representation of summary statistics using box plots, while Figure 9.2 shows the proportion of results belonging to different air quality categories.
Table 4: Example of table showing PM10 summary statistics
PM10 (µg/m3) at (airshed name)
|Monitoring site: (physical address of site) |
Classification: Residential – neighbourhood
From 1 January 2008 to 31 December 2008
24-hour concentrations (ending at midnight each day)
Per cent valid data: 96% Data capture rate: 100%
|Number of exceedences||0||0||0||0||0||2||5||8||3||0||0||0|
Source: Auckland Regional Council.
Figure 9.2: Example of figure showing the proportion of PM10 samples belonging to different air quality categories