Article
BAMOS November 2025
19
Density of the network and site spacing
The network was rolled out in three phases. The first phase covered much of the Midnorth( the region centred around the Clare Valley north of Adelaide). This region is complex in topography, landscape and nearby coastal configuration. The Air Pollution Model( TAPM) from CSIRO, a mesoscale model( Hurley, 2005), was run at 1 km resolution for five model years, focusing on minimum temperatures and turbulence kinetic energy as proxies for stability. Composites of events for extreme minima identified cold-air catchments and drainage flows, which were matched with local experience. From this, discrete coherent regions were identified as potential sites. Although the focus is on spray drift, a similar procedure for fire danger rating also informed the outcome.
Some sites, such as regional agricultural research centres, were predetermined. In addition, Mt Bryan, a bare, sharp peak at 935 m altitude and the highest point in the Mount Lofty Ranges, was selected as it represents conditions at the gradient wind level, making it useful for nowcasting. For this region, the median nearest-neighbour distance of stations is 12 to 15 km( compared to 20 km for the remainder).
For the Riverland & Mallee and the Limestone Coast( second and third phases, respectively), site spacing was determined by combining the accumulated empirical knowledge of spatial variability with analyses of the Bureau of Meteorology’ s Atmospheric high-resolution Regional Reanalysis for Australia( BARRA) dataset, which was available albeit at ~ 11 km resolution( Su et al., 2019). Siting was further adjusted to account for void areas due to the absence of agriculture, as well as a desired level of redundancy and proximity to BOM stations for data comparison.
A wide range of sensors and measurements
The sensor suite for each station( Table 1) is broader than that of typical weather stations. Two sonic anemometers and a vertical temperature difference between 1.2 and 10 m support routine micrometeorological measurements( Figure 2). Measurements are performed over 10-minute periods with 4 Hz sampling to ensure accurate variances and covariances. The primary temperature sensor is accurate to ± 0.6 ° C, which compares to BOM stations’ sensor accuracy of ± 0.3 ° C. For VTD, paired thermocouples provide accuracy of ± 0.2 ° C. The integration of sensors, data logging, communications, antennas, and power is provided by Campbell Scientific Australia, although some legacy systems still operate.
Ensuring data quality a) Procedures
Physical inspection and maintenance of the site and equipment is performed annually and reactively as required. Site selection, maintenance and quality assurance procedures closely follow US Mesonets, especially the Oklahoma Mesonet( McPherson et al., 2007, and Fiebrich et al., 2010), which in turn are compatible with the standards of the World Meteorological Organization( 2021-23). Metadata is maintained and comprises all quality assurance and control reports and a set of station dossiers with site details and photographic records. Standards for accuracy, timeliness and availability continue to mature as stretch goals.
Focusing on accuracy, there are four main time frames for quality assurance: daily, opportunistic‘ archetype’ events, monthly and long-term. Shortly after midnight, all daily observations are checked for receipt, delays, domain consistency, internal logic, and spatial and temporal coherence, and an email summary is distributed. An example of an archetypal event is a period
Table 1: Instrumentation on a typical Mesonet station. All anemometers at 10 m are to be 3D versions as resources permit. A tradecraft feature of the Mesonet towers allows VTD to be field-calibrated to zero. Some stations have legacy equipment.
Figure 2: The suite of instruments on a typical Mesonet station. Credit: Warwick Grace
Hurley, P. J. 2005. The Air Pollution Model( TAPM) Version 3. Part 1: Technical description. CSIRO Atmospheric Research Technical Paper No. 71, CSIRO. Su, C H., Eizenberg, N., Steinle, P., Jakob, D., Fox Hughes, P., White, C. J., Rennie, S., Franklin, C., Dharssi, I. and Zhu, H. 2019. FAQ for BARRA( Bureau of Meteorology Atmospheric high resolution Regional Reanalysis for Australia). Bureau of Meteorology.