Blair Trewin's epic journey to 112 Australian weather stations

Blair Trewin is a wonderful character and one of the leading researchers of the homogenization community. He works at the Australian Bureau of Meteorology (BOM) and created their high-quality homogenized datasets. He also developed a correction method for daily temperature observations that is probably the best we currently have. Fitting to his scientific love of homogenization, he has gone on a quest to visit all 112 weather stations that are used to monitor the Australian climate. Enjoy the BOM blog post on this "epic journey".

To Bourke and beyond: one scientist’s epic journey to 112 weather stations

There are 112 weather observation stations that feed into Australia’s official long-term temperature record—and Bureau scientist, Blair Trewin, has made it his personal mission to visit all of them! Having travelled extensively across Australia—from Horn Island in the north to Cape Bruny in the south, Cape Moreton in the east to Carnarvon in the west—Blair has now ticked off all but 11 of those sites.

---

Map: the 112 observation locations that make up Australia's climate monitoring network

Some of the locations are in or near the major cities, but many are in relatively remote areas and can be difficult to access. Blair says perhaps his most adventurous site visit was on the 2009 trip at Kalumburu, an Aboriginal community on the northernmost tip of the Kimberley, and two days’ drive on a rough track from Broome. ‘I asked the locals the wrong question—they said I’d be able to get in, but I didn’t ask them whether I could get back out again’. After striking trouble at a creek crossing leaving town, he spent an unplanned week there waiting for his vehicle to be put on a barge back to Darwin.

While these locations are remote now, in some ways they were even more remote in the past. These days you can get a signal for your mobile phone in Birdsville, Queensland, but as recently as the 1980s, the only means of rapid communication was often-temperamental radio relays through the Royal Flying Doctor Service. Today distance is no longer an issue; the majority of weather stations in the Bureau’s climate monitoring network—including Birdsville—are automated, with thermometers that submit the information electronically.

Photo: Blair Trewin at the weather observation station at Tarcoola, in the far north of South Australia. The Stevenson screen houses a resistance temperature device (thermometer) and a relative humidity probe

But, even some of the sites closer to home have posed a challenge for Blair’s mission. To get to Gabo Island in Victoria for example, you need to either fly or take a boat, and the runway is just a few hundred metres long, so it can only be used in light winds. ‘I spent two days in Mallacoota waiting for the winds to drop enough to get over there’.

Similarly, the site at the Wilsons Promontory lighthouse, if you don’t use a helicopter, is accessed through a 37 km return hike, which Blair did as a training run with one of his Victorian orienteering teammates.

You can read the rest of this adventure at the Blog of the Australian Bureau of Meteorology.

The transition to automatic weather stations. We’d better study it now.

This is a POST post.

The Parallel Observations Science Team (POST) is looking across the world for climate records which simultaneously measure temperature, precipitation and other climate variables with a conventional sensor (for example, a thermometer) and modern automatic equipment. You may wonder why we take the painstaking effort of locating and studying these records. The answer is easy: the transition from manual to automated records has an effect on climate series and the analysis we do over them.

In the last decades we have seen a major transition of the climate monitoring networks from conventional manual observations to automatic weather stations. It is recommended to compare these instruments before the substitution is effective with side by side measurements, which we call parallel measurements. Climatologists have also set up many longer experimental parallel measurements. They tell us that in most cases both sensors do not measure the same temperature or collect the same amount of precipitation. A different temperature is not only due to the change of the sensor itself, but automatic weather stations also often use a different, much smaller, screen to protect the sensor from the sun and the weather. Often the introduction of automatic weather stations is accompanied by a change in location and siting quality.

From studies of single temperature networks that made such a transition we know that it can cause large jumps; the observed temperatures at a station can go up or down by as much as 1°C. Thus potentially this transition can bias temperature trends considerably. We are now trying to build a global dataset with parallel measurements to be able to quantify how much the transition to automatic weather stations influences the global mean temperature estimates used to study global warming.

Temperature

This study is led by Enric Aguilar and the preliminary results below were presented at the Data Management Workshop in Saint Gallen, Switzerland last November. We are still in the process of building up our dataset. Up to now we have data from 10 countries: Argentina (9 pairs), Australia (13), Brazil (4), Israel (5), Kyrgyzstan (1), Peru (31), Slovenia (3), Spain (46), Sweden (8), USA (6); see map below.


Global map in which we only display the 10 countries for which we have data. The left map is for the maximum temperature (TX) and the right for the minimum temperature (TN). Blue dots mean that the automatic weather station (AWS) measures cooler temperatures than the conventional observation, red dots mean the AWS is warmer. The size indicates how large the difference is, open circles are for statistically not significant differences.

The impact of the automation can be better assessed in the box plots below.


The bias of the individual pairs are shown as dots and summarized per country with box plots. For countries with only a few pairs the boxplots should be taken with a grain of salt. Negative values mean that the automatic weather stations are cooler. We have data for Argentina (AR), Australia (AU), Brazil (BR), Spain (ES), Israel (IL), Kyrgyzstan (KG), Peru (PE), Sweden (SE), Slovenia (SI) and the USA (US). Panels show the maximum temperature (TX), minimum temperature (TN), mean temperature (TM) and Diurnal temperature range (DTR, TX-TN).

On average there are no real biases in this dataset. However, if you remove Peru (PE) the differences in the mean temperature are either small or negative. That one country is so important shows that our dataset is currently too small.

To interpret the results we need to look at the main causes for the differences. Important reasons are that Stevenson screens can heat up in the sun on calm days, while automatic sensors are sometimes ventilated. The automatic sensors are, furthermore, typically smaller and thus less affected by direct radiation hitting them than thermometers. On the other hand, in case of conventional observation, the maintenance of the Stevenson screens—cleaning and painting—and detection of other problems may be easier because they have to be visited daily. There are concerns that plastic screens get more grey and heat more in the sun. Stevenson screens have more thermal inertia, they smooth fast temperature fluctuations, and will thus show lower highs and higher lows.

Also the location often changes with the installation of automatic weather stations. America was one of the early adopters. The US National Weather Service installed analogue semi-automatic equipment (MMTS) that did not allow for long cables between the sensor and the display inside a building. Furthermore, the technicians only had one day per station and as a consequence many of the MMTS systems were badly sited. Nowadays technology has advanced a lot and made it easier to find good sites for weather stations. This is maybe even easier now than it used to be for manual observations because modern communication is digital and if necessary uses radio making distance much less a concern. The instruments can be powered by batteries, solar or wind, which frees them from the electricity grid. Some instruments store years of data and need just batteries.

In the analysis we thus need to consider whether the automatic sensors are placed in Stevenson screens and whether the automatic weather station is at the same location. Where the screen and the location did not change (Israel and Slovenia), the temperature jumps are small. Whether the automatic weather station reduces radiation errors by mechanical ventilation is likely also important. Because of these different categories, the number of datasets needed to get a good global estimate becomes larger. Up to now, these factors seem to be more important than the climate.

Precipitation

For most of these countries we also have parallel measurements for precipitation. The figure below was made by Petr Stepanek, who leads this part of the study.


Boxplots for the differences in monthly precipitation sums due to automation. Positive values mean that the manual observations record more precipitation. Countries are: Argentina (AG), Brazil (BR), The Check Republic (CZ), Israel (IS), Kyrgyzstan (KG), Peru (PE), Sweden (SN), Spain (SP) and the USA (US). The width of the boxplots corresponds to the size of the given dataset.

For most countries the automatic weather stations record less precipitation. This is mainly due to smaller amounts of snow during the winter. Observers often put a snow cross in the gauge in winter to make it harder for snow to blow out of it again. Observers simply melt the snow gathered in a pot to measure precipitation, while early automatic weather stations did not work well with snow and sticky snow piling up in the gauge may not be noticed. These problems can be solved by heating the gauge, but unfortunately the heater can also increase the amount of precipitation that evaporates before it could be registered. Such problems are known and more modern rain gauges use different designs and likely have a smaller bias again.

Database with parallel data

The above results are very preliminary, but we wanted to show the promise of a global dataset with parallel data to study biases in the climate record due to changes in the observing practises. To proceed we need more datasets and better information on how the measurements were performed to make this study more solid.

In future we also want to look more at how the variability around the mean is changing. We expect that changes in monitoring practices have a strong influence on the tails of the distribution and thus on estimates of changes in extreme weather. Parallel data offer a unique opportunity to study this otherwise hard problem.

Most of the current data comes from Europe and South America. If you know of any parallel datasets especially from Africa or Asia, please let us know. Up to now, the main difficulty for this study is to find the persons who know where the data is. Fortunately, data policies do not seem to be a problem. Parallel data is mostly seen as experimental data. In some cases we “only” got a few years of data from a longer dataset, which would otherwise be seen as operational data.

We would like to publish the dataset after publishing our papers about it. Again this does not seem to lead to larger problems; sometimes people prefer to first publish an article themselves, which causes some delays, and sometimes we cannot publish the daily data itself, but “only” monthly averages and extreme value indices, this makes the results less transparent, but these summary values contain most of the information.

Knowledge of the observing practices is very important in the analysis. Thus everyone who contributes data is invited to help in the analysis of the data and co-author our first paper(s). Our studies are focused on global results, but we will also provide everyone with results for their own dataset to gain a better insight into their data.

Most climate scientists would agree that it is important to understand the impact of automation on our records. So does the World Meteorological Organization. In case it helps you to convince your boss: the Parallel Observations Science Team is part of the International Surface Temperature Initiative (ISTI). It is endorsed by the Task Team on Homogenization (TT-HOM) of the World Meteorological Organization (WMO).

We expect that this endorsement and our efforts to raise awareness about our goals and their importance will help us to locate and study parallel observations from other parts of the world, especially Africa and Asia. We also expect to be able to get more data from Europe; the regional association for Europe of the WMO has designated the transition to automatic weather stations as one of its priorities and is helping us to get access to more data. We want to have datasets for all over the world to be able to assess whether the station settings (sensors, screens, data quality, etc.) have an impact, but also to understand if different climates produce different biases.

If you would like to collaborate or have information, please contact me.

Tweet

Related reading

The ISTI has made a series of brochures on POST in English, Spanish, French and German. If anyone is able to make further translations, that would be highly appreciated.

Parallel Observations Science Team of the International Surface Temperature Initiative.

Irrigation and paint as reasons for a cooling bias

Temperature trend biases due to urbanization and siting quality changes

Changes in screen design leading to temperature trend biases

Temperature bias from the village heat island

Changes in screen design leading to temperature trend biases

In the lab, temperature can be measured with amazing accuracies. Outside, exposed to the elements, measuring the temperature of the air is much harder. For example, if the temperature sensor gets wet, due to rain or dew, the evaporation leads to a cooling of the sensor. The largest cause of exposure errors are solar and heat radiation. For these reasons, thermometers need to be protected from the elements by a screen. Changes in the radiation error are an important source of non-climatic changes in station temperature data. Innovations leading to reductions in these errors are an major source of temperature trend biases.

A wall measurement at the Mathematical Tower in Kremsmünster. You mainly see the bright board to protect the instruments against rain, which is on the first floor, at the base of the window, a little right of the entrance.

History

The history of changes in exposure is different in every country, but in broad lines follows this pattern. In the beginning thermometers were installed in unheated rooms or in front of a window of an unheated room on the North (poleward) side of a building.

When this was found to lead to too high temperatures a period of innovation and diversity started. For example, small metal cages were added to the North wall measurements. More importantly free standing structures were designed: stands, shelters, houses and screens. In the Common Wealth the Glaisher (Greenwich) stand was prevalent. It has a vertical wooden board, a small roof and sides, but it is fully open in the front and in summer you have to rotate it to ensure that no direct sun gets onto the thermometer.

Shelters were build with larger roofs and sides, but still open to the front and the bottom, for example the Mountsouris and Wild screens. Sometimes even small houses or garden sheds were build, in the tropics with a thick thatched roof.

In the end, the [[Stevenson screen]] (Cotton Region Shelter) won the day. This screen is closed to all sides. It has double Louvre walls, double boards as roof and a board as bottom. (Early designs sometimes did not have a bottom.)

In the recent decades there is a move to Automatic Weather Stations (AWS), which do not have a normal (liquid in glass) thermometer, but an electrical resistance temperature sensor and is typically screened by multiple round plastic cones. These instruments are sometimes mechanically ventilated, reducing radiation errors during calm weather. Some countries have installed their automatic sensors in Stevenson screens to reduce the non-climatic change.

The photo on the left shows an open shelter for meteorological instruments at the edge of the school square of the primary school of La Rochelle, in 1910. On the right one sees the current situation, a Stevenson-like screen located closer to the ocean, along the Atlantic shore, in place named "Le bout blanc". Picture: Olivier Mestre, Meteo France, Toulouse, France.

Radiation errors

To understand when and where the temperature measurements have most bias, we need to understand how solar and heat radiation leads to measurement errors.

The temperature sensor should have the temperature of the air and should thus not be warmed or cooled by solar or heat radiation. The energy exchange between sensor and air due to ventilation should thus be large relative to the radiative exchanges. One of the reasons why temperature measurements outside are so difficult is that these are conflicting requirements: closing the screen for radiation will also limit the air flow. However, with a smart design, mechanical ventilation and small sensors this conflict can be partially resolved.

For North-wall observations direct solar radiation on the sensor was sometimes a problem during sunrise and sunset. In addition the sun may heat the wall below the thermometer and warm the rising air. Even for Stevenson screens some solar radiation still gets into the screen. Furthermore, the sun shining on the screen warms it, which can then warm the air flowing through the screen. For this reason it is important that the screen is regularly painted white and cleaned.

Scattered solar radiation (clouds, vegetation, surface) is important for older screens being open to the front. The open front also leads to a direct cooling of the sensor as it emits heat radiation. The net heat radiation flux is especially large when the back radiation of the atmosphere is low, thus when there are no clouds and the air is dry. Warm air can contain more humidity, thus these effects are generally also largest when it is cold.

Because older screens did not have a bottom, a hot surface below the screen could be a problem during the day and a cold surface during the night. This especially happens when the soil is dry and bare.

All these effects are most clearly seen when the wind is calm.

Concluding, we expect the cooling bias at night to be largest when the weather is calm, cloud free and the air is dry (cold). We also expect a warming bias during the day to be largest when the weather is calm and cloud free. In addition we can get a warm bias when the soil is dry and bare and in summer during sunrise and sunset.

Thus all things being equal, the radiation error is expected to be largest in sub-tropic, tropical and continental climates and small in maritime, moderate and cold climates.


Schematic drawing of the various factors that can lead to radiation errors.

Parallel measurements

We know how large these effects are from parallel measurements, where an old and new measurement set-up are compared side by side. Unfortunately, there are not that many of parallel measurements for the transition to Stevenson screens. Many parallel measurements in North-West Europe, a maritime, moderate or cold climate, where the effects are expected to be small of those are described in a wonderful review article by David Parker (1994) and he concludes that in the mid-latitudes the past warm bias will be smaller than 0.2°C. In the following, I will have a look at the parallel measurements outside of this region.

In the topics, the bias can be larger. Parker also describes two parallel measurements of a tropical thatched house with a Stevenson screen. One in India and one in Ceylon (Sri Lanka). They both have a bias of about 0.4°C. The bias naturally depends on the design, a comparison of a normal Stevenson screen with one with a thatched roof in Samoa shows almost no differences.

This picture shows three meteorological shelters next to each other in Murcia (Spain). The rightmost shelter is a replica of the Montsouri (French) screen, in use in Spain and many European countries in the late 19th century and early 20th century. In the middle, Stevenson screen equipped with automatic sensors. Leftmost, Stevenson screen equipped with conventional meteorological instruments. Picture: Project SCREEN, Center for Climate Change, Universitat Rovira i Virgili, Spain.

Recently two beautiful studies were made with modern automatic equipment to study the influence of the screens. With automatic sensors you can make measurements every 10 minutes, which helps in understanding the reasons for the differences. In Spain they have build two replicas of the French screen used around 1900. One was installed in [[La Coruna]] (more Atlantic) and one in [[Murcia]] (more Mediterranean). They showed that the old measurements had a temperature bias of about 0.3°C; the Mediterranean location had, as expected, a somewhat larger bias than the Atlantic one.

The second modern study was in Austria, at the Mathematical Tower in Kremsmünster (depicted at the top of this post). This North-wall measurement was compared to a Stevenson screen (Böhm et al., 2010). It showed a temperature bias of about 0.2°C. The wall was oriented North-North-East and during sunrise in summer the sun could shine on the instrument.

For both the Spanish and the Austrian examples it should be noted that small modern sensors were used. It is possible that the radiation errors would have been larger had the original thermometers been used.

Comparing a Wild screen with a Stevenson screen at the astronomical observatory in [[Basel]], Switzerland, Renate Auchmann and Stefan Brönnimann (2012) found clear signs of radiation errors, but the annual mean temperature was somehow not biased.

Parallel measurement with a Wild screen and a Stevenson screen in Basel, Switzerland.

In [[Adelaide]], Australia, we have a beautiful long parallel measurement of the Glaisher (Greenwich) stand with a Stevenson screen (Cotton Region Shelter). It runs 61 complete years (1887-1947) and shows that the historical Glaisher stand recorded on average 0.2°C higher temperatures; see figure with annual cycle below. The negative bias in the minimum temperature at night is almost constant throughout the year, the positive bias is larger and strongest in summer. Radiation errors thus not only affect the mean, but also the size of the annual cycles. They will also affect the daily cycle, as well as the weather variability and extremes in the temperature record.

The exact size of the bias of this parallel measurement has a large uncertainty, it varies considerably from year to year and the data also shows clear inhomogeneities itself. For such old measurements, the exact measurement conditions are hard to ascertain.

The annual cycle of the temperature difference between a Glaisher stand and a Stevenson screen. For both the daily maximum and the daily minimum temperature. (Figure 1 from Nicholls et al. (1996)

Conclusions

Our understanding of the measurements and limited evidence from parallel measurements suggest that there is a bias of a few tenth of a Centigrade in observations made before the introduction of Stevenson screens. The [[Stevenson screen]]
was designed in 1864, most countries switched in the decades around 1900, but some countries did not switch until the 1960ies.

The last few decades there was a new transition to automatic weather stations (AWS). Some countries have installed the automatic probes in Stevenson screens, but most have installed single unit AWS with multiple plastic cones as screen. The smaller probe and mechanical ventilation could make the radiation errors smaller, but depending on the design possibly also more radiation gets into the screen and the maintenance may also be worse now that the instrument is no longer visited daily. An review article on this topic is still dearly missing.

Last month we have founded the Parallel Observations Science Team (POST) as part of the International Surface Temperature Initiative (ISTI) to gather and analyze parallel measurements and see how they affect the climate record. (Not only with respect to the mean, but also for changes in day and annual cycles, weather variability and weather extremes.) Theo Brandsma will lead our study on the transition to Stevenson screens and Enric Aguilar the transition from conventional observations to automatic weather stations. If you know of any dataset and/or want to collaborate please contact us.

Acknowledgement

With some colleagues I am working on a review paper on inhomogeneities in the distribution of daily data. This work, especially with Renate Auchmann, has greatly helped me understand radiation errors. Mistakes in this post are naturally my own. More on non-climatic changes in daily data later.

Tweet

Further reading

A beautiful "must-read" article on temperature screens by Stephen Burt: What do we mean by ‘air temperature’? Measuring temperature is not as easy as you may think.

Just the facts, homogenization adjustments reduce global warming: The adjustments to the land surface temperature increase the trend, but the adjustments to the sea surface temperature decrease the trend.

Temperature bias from the village heat island

A database with parallel climate measurements describes the database we want to build with parallel measurements

A database with daily climate data for more reliable studies of changes in extreme weather gives somewhat more background

Statistical homogenisation for dummies

New article: Benchmarking homogenisation algorithms for monthly data

References

Auchmann, R., and S. Brönnimann, 2012: A physics-based correction model for homogenizing sub-daily temperature series. Journal Geophysical Research, 117, D17119, doi: 10.1029/2012JD018067.

Böhm, R., P.D. Jones, J. Hiebl, D. Frank, M. Brunetti, M.Maugeri, 2010: The early instrumental warm-bias: a solution for long central European temperature series 1760–2007. Climatic Change, 101, no. 1-2, pp 41-67, doi: 10.1007/s10584-009-9649-4.

Brunet, M., Asin, J., Sigró, J., Bañón, M., García, F., Aguilar, E., Palenzuela, J. E., Peterson, T. C. and Jones, P., 2011: The minimization of the screen bias from ancient Western Mediterranean air temperature records: an exploratory statistical analysis. International Journal Climatology, 31, pp, 1879–1895, doi:
10.1002/joc.2192.

Nicholls, N., R. Tapp, K. Burrows, D. Richards. Historical thermometer exposures in Australia. International Journal of Climatology, 16, pp. 705-710, doi: 10.1002/(SICI)1097-0088(199606)16:6<705::AID-JOC30>3.0.CO;2-S, 1996.

Parker, D. E., 1994: Effects of changing exposure of thermometers at land stations. International Journal of Climatology, 14, pp. 1–31, doi: 10.1002/joc.3370140102.