Automatic weather station influence on Cape Bruny temperature trendsAustralia ACORN 2 analysis New South Wales Northern Territory Queensland South Australia Tasmania Victoria Western Australia
In the absence of any other influences, an instrument with a faster response time will tend to record higher maximum and lower minimum temperatures than an instrument with a slower response time. This is most clearly manifested as an increase in the mean diurnal range. At most locations (particularly in arid regions), it will also result in a slight increase in mean temperatures, as short-term fluctuations of temperature are generally larger during the day than overnight (Trewin, 2018). ... To assess the effective response time of the observations, mean values of the one-minute temperature variation (that is, the difference between the highest and lowest temperature in a 1-minute period) were calculated from the available one-minute data. At 1500 local time in summer (near the typical time of maximum temperature), these mean values are typically between 0.15 °C and 0.30 °C, reaching up to 0.40 °C in some arid regions, but at 1500 in winter they are below 0.10 °C at most southern locations. At 0600 local time, near the usual time of minimum temperature, mean values of the one-minute variation are between 0.05 °C and 0.10 °C all year. At 17 of the 98 ACORN-SAT locations with automatic weather stations and one-minute data, there was a significant breakpoint during the period of automatic observations in the time series of annual means of the mean one-minute temperature variation. Such a breakpoint indicates a likely change in the response time of the instrument, and will hence affect its sampling of maximum and minimum temperatures. In 16 of the 17 cases, the breakpoint coincided with the documented replacement of a Rosemount probe with no version number (some were replaced by a Rosemount ST2401, some by a Wika TR40 and some by a Temp Control). Figure 8 shows two examples of this. Alice Springs is the most extreme example; the November 2011 probe replacement there resulted in an increase in mean one-minute temperature fluctuations of approximately 0.16 °C at 1500 and 0.03 °C at 0600. Assuming that the increased variation is distributed symmetrically about the one-minute mean and that the change of probe did not introduce any inhomogeneities into the one-minute mean, this equates to an upward shift of about 0.08 °C for maximum temperature and a downward shift of 0.01-0.02 °C in minimum temperature. In less arid climates the effect is smaller (e.g., for Sydney, around +0.03 °C for maxima and –0.01 °C for minima). Given the relatively small proportion of the network which is affected, it is estimated that the overall effect on national maximum temperatures is in the order of +0.01 °C, and on minimum temperature, between zero and −0.01 °C. In the context of overall Australian temperature change and variability, these network-wide impacts are negligible, whilst even at the worst-affected stations, the size of the impact falls well below the 0.3 °C minimum threshold normally applied for station-specific adjustments in the ACORN-SAT dataset. No specific adjustment for this change was therefore made in version 2 of ACORN-SAT. Below is the description of Tasmania's Cape Bruny weather station extracted from the ACORN-SAT Station Catalogue. The site appears to have been close to its current location before a move in April 1939. It also moved on 13 November 1961, and 45 m east in March 1969. An automatic weather station (094198) has been operating since 1997 at Cape Bruny. This is expected eventually to supersede the current site. It is about 50 m west of the current site, close to the 1961–1969 location of the manual site, but is in a more exposed location on top of the ridge. This exposure has clearly reduced rainfall readings but impacts on temperature are less clear. In other words, Cape Bruny provides a fairly unique comparison of AWS vs manual Stevenson screens just 50 metres apart over 20 years of observation. The ACORN 2 technical report acknowledges the influence of response times on maxima and minima stemming from a change in AWS probes but doesn't acknowledge the potential difference with manual liquid thermometers which have a slower response time to changes in air temperature.
94010 manual 1998-2017 average max 15.57C vs 94198 AWS 1998-2017 (50 metres west) average max 16.02C 94010 manual 1998-2017 average min 9.47C vs 94198 AWS 1998-2017 (50 metres west) average min 9.28C 94010 has an elevation of 55 metres and 94198 has an elevation of 60 metres, so there’s a five metre upward slope west of 94010 over 50 metres. With the sun setting to the west, the higher slope should only start to cast shadow on the lower screen very late in the afternoon, a few hours after the daily maximum has been recorded. 94010 has had an average rainfall of 879.0mm since 1998 and 94198 has had an average annual rainfall of 763.5mm since 1998. Bureau of Meteorology researcher Blair Trewin explains this marked difference in rainfall just 50 metres apart here. "One example is at Cape Bruny in Tasmania, where we have two sites 50 m apart, one on the top of a hill and one just off to the side in a more sheltered location. Visiting there on a windy, showery, day made me appreciate how much more exposed the hilltop site is, and it turns out the hilltop site measures 17 per cent less rain than the more sheltered one, because some of the rain blows over the top of the gauge." The 94198 AWS averages a 0.45C warmer maximum per annum than the manual 94010 screen. 94010 is cooler than the AWS each month by ... Jan -0.8C Either the AWS response time is more of an influence during warmer weather, as effectively acknowledged by the bureau, or something else about the stations' environment is reducing the difference as the months get cooler (and the days get shorter). Note the description in the ACORN Catalogue about the position of 94198 ... "It is about 50 m west of the current site, close to the 1961–1969 location of the manual site ..." In the eight available annual observations during 1961-1969, the Cape Bruny 94010 manual site average raw maximum was 14.88C. When shifted east to its current site, the eight years of 1970-1977 had an annual average raw maximum of 15.30C (even though average annual rainfall was 950.6mm in 1961-69 and 1,098.2mm in 1970-77). The homogenised ACORN 1 dataset shows that Tasmania's overall maximum temperature increased 0.21C from 1961-1969 to 1970-1977. The closest ACORN station, Grove (57 kilometres distant), had an average maximum of 16.83C in 1961-1969 and 16.92C in 1970-1977 - an increase of 0.09C. This means Cape Bruny's manual screen was 0.42C cooler when in the western location where the current AWS is now recording 0.45C warmer. The higher 94198 location should also be more exposed to cooler westerly, easterly and southerly winds off the ocean - powerful enough to blow 17% of rainfall over the gauges, according to Trewin. The 94198 annual mean 3pm wind speed is 25.7kmh and the 94010 annual mean 3pm wind speed is 27.8kmh. Most Cape Bruny winds are westerlies and southerlies. This suggests the 94198 AWS location isn’t a warmer environment and the warmer temperature readings are more likely due to instrumental influences such as the more rapid response time of the AWS. It is generally believed that the rapid response warming of maxima by Automatic Weather Stations becomes more pronounced as the temperature increases. The table below compares the difference in the warmer and cooler years recorded at Cape Bruny's 94198 AWS and 94010 manual screens from 1998 to 2017.
Cape Bruny is the most southerly of all ACORN stations and, as a result, one of the coldest near sea level. The rainfall shows it’s not arid. 94010 is surrounded by grass and 94198 by heath to 0.5m, but presumably well exposed to avoid vegetation shade. Maximum temperatures are normally recorded mid afternoon when shading isn’t an issue. Stevenson screens are not supposed to be located on watered grass but it’s unlikely the grass pictured above changes colour or heat radiation due to seasonal drying as there is very little difference in Cape Bruny’s rainfall between the 12 months, and all months are rainy enough to maintain the green. The driest month is February with an average 56.12mm and the wettest is July with an average 95.1mm, meaning in the driest month the location still gets an average 2mm of rainfall every day. This should also maintain the greenery and albedo of the surrounding heath and 3m shrubs just east of the AWS. The bureau's ACORN 2 technical report explanation of AWS response times minimises artificial warming of maximum temperatures by confining it only to a small number of hot, arid locations in Australia and thus not sufficiently influential to adjust the national homogenised dataset. Cape Bruny poses questions not only about the AWS response time warming at the Tasmanian location itself but the response warming at all Australian sites - hot, cold, arid or wet.
|
|
Home |
