About Weather Station Data

Introduction

This page provides some information about the Bureau's observation of the weather which may help you selecting the type of data you require. It briefly provides information about where you may find a Bureau weather station, what observations may have been made and why they differ between stations, and why there can be a different number of observations available within a day.

Location of stations

Weather data are obtained from different types of observing stations around Australia, on offshore islands, and in the Antarctic. They include Bureau-staffed and cooperative observer stations, automatic weather stations, meteorological satellites and drifting ocean buoys, ships and aircraft that collect weather data. Station locations are chosen based on a range of different requirements (PDF, 600 kB) and constraints, including:

  • The need for observations of particular types;
  • How well a location represents the surroundings;
  • Availability of land, observers (if required) and infrastructure;
  • Historical decisions;
  • Other stations in the area.

Sometimes the closest station may not be the best selection. For example, if you are interested in historical climate data in a valley at the bottom of a steep mountain range, and the nearest weather station is on the top of the range, a more appropriate data set may be from another station also in the valley, despite it being a little further away.

top

Observations available from a station

There are many weather elements and associated parameters which may be recorded at a weather station. What a station reports can change over the period it is in operation. For example, a station may start out being a staffed station which reports on a wide range of weather phenomena, but due to changing requirements, may be changed to a rainfall only station. In using the Weather Station Directory it can be beneficial to look at the data availability table and review the period over which a particular element has been observed, and how complete the record is. For example, a station may measure wind speed at the time it commenced operation, this may stop for several decades, and start again as operation requirements change.

top

Reporting frequencies and times of observation

The monthly, daily, 3 hourly, half hourly, and minute frequency options represent the most typical reporting schedules within broader reporting ranges.

  • Minute: One-minute observations are recorded every one minute.
  • Half Hourly: Half hourly observations are recorded every half hour, or at non-routine times (known as special observations, which are generated when variations in weather conditions exceed certain thresholds).
  • 3 Hourly: 3 hourly observations (also known as synoptic or hourly) are observations recorded at 3 hourly time intervals. 3 hourly observations may only be reported at 9 am and 3 pm at some stations; at 6 am, 9 am and 3 pm at others; or every 3 hours.
  • Daily: Daily observations are recorded over a 24 hour period. Some variables, such as maximum wind gust, hours of bright sunshine, and weather phenomena (snow, fog, hail, dust, frost, thunder) are measured from midnight to midnight, whilst most other daily elements are measured over the 9 am to 9 am period. More information. Whether the 9am observation is attributed to the day of observation or that preceding depends on the element being recorded:
    • Rainfall, 24 total recorded against the day of observation.
    • Evaporation, 24 total recorded against the day of observation.
    • Maximum temperature, highest value recorded against the previous day.
    • Minimum temperature, lowest value recorded against the day of observation.
    • Daily wind run, 24 total recorded against the day of observation.
    On some occasions, perhaps through an observer being sick, daily observations are accumulated over more than 24 hours. Data are still assigned to dates as above, but the period of accumulation (if available) is generally provided along with the data.
  • Monthly: In addition to the above reporting frequencies, the Bureau also provides monthly data. These are typically monthly averages (e.g. temperature) or totals (e.g. monthly rainfall), and are derived from hourly and/or daily observations.

Why are there different reporting frequencies?

At the majority of locations, Automatic Weather Stations (AWSs) send data frequently. Some provide data every minute, while others report on an hourly basis. The AWSs are designed to provide data for the Bureau's forecasting, warning, and information services, as well as providing data for the Bureau's climate database.

At some locations Bureau observers supplement the temperature, humidity, pressure and wind observations with observations of cloud and other elements such as sea state. These are referred to as 'Visual Observations' in some products. These manual observations are reported less frequently: sometimes only at 9 am, at 9 am and 3 pm, or every 3 hours at some stations. For most of the day, three hourly instrument readings are made of temperature, air pressure, humidity, rainfall, and wind speed and direction, and visual observations are made of cloud cover and visibility. At some stations the temperature and humidity of the upper atmosphere are also obtained by balloon-borne instruments to an altitude of 25 kilometres.

top

Station metadata reports for information about stations and available observations

Detailed site information is provided in PDF format for each weather station. You can find these PDFs either by clicking the "More details" link under a station name when presented with a list of available station in the Weather Station Directory, or by clicking the map icon with "Details" above the table of data in an observations page in Climate Data Online.

Example of metadata PDF location on weather station directory

Where to find metadata PDFs on the Weather Station Directory

Example of metadata PDF location on climate data online pages

Where to find metadata PDFs on Climate Data Online pages

The metadata report, which is updated periodically, provides basic climate metadata (information about data) for individual sites. Notes about the metadata are included in the PDF file. Extended information, including equipment listings, is also available by contacting us. Generally, station summaries which are mentioned on page one of the document are not yet available.

The tables of data holdings - starting on page two - are useful for quickly determining the periods for which data are available at a site, including finding years where some or all observations are missing. It summarises the observations available in the Australian Data Archive for Meteorology (ADAM) at the time the information was compiled, and provides information additional to the site lists available from the Weather station network lists, but on a site by site basis.

Example of data holdings diagram

The circled numbers in the sample data holdings image above indicate:

  1. The climate element being summarised.
  2. The month and year of the first observation. This might not be the same for all elements.
  3. The month and year of the last observation. For open stations still observing this element, the value will be the date the information was compiled.
  4. An estimate of how complete the record is, between the first and last dates.
  5. The number of days, excluding complete months, of missing data.
  6. The number of complete months without data (for that element).
  7. The colour code which represents data completeness on a yearly basis.
  8. The year-by-year availability of data, with dividers representing the start of the decade.

The data holdings for synoptic observations (observations within a day used for weather forecasting) has an additional field, which provides the average frequency of observations per day, calculated over the complete period of record. Data holdings for half-hourly and one minute data have not been sub-divided into their individual elements, but represent the observation report in total. An individual element may be missing, for example, if there was an equipment malfunction.

top