138: Evidence against temperature adjustments #3 (Scandinavia)

One of the main aims of this blog has been to investigate the extent to which the various datasets in the global temperature record have been adjusted and to ascertain both the impact of these adjustments and their validity. Most of the blog posts for individual countries or territories have sought to quantify the magnitude of these adjustments by calculating two versions of the mean temperature anomaly (MTA) for each region; one based on its raw unadjusted data and a second using Berkeley Earth adjusted data. Then the two are compared and the difference calculated. This difference is often considerable and often shows that the adjustments have increased the amount of reported warming. But I have also investigated the second issue, that of validity. One way to do this is to compare the MTA for neighbouring regions or different data samples from the same region. 

The rationale is as follows. If there are errors in the data that are sufficient to affect the MTA, then comparing MTAs from different samples from the same region, or samples from adjacent regions that would be expected to be almost identical, could highlight the errors. Of course any difference between MTAs from different regions does not prove that the data is wrong; it may be that the regions aren't as similar as one supposed. But if the data is virtually identical then that does suggest both that the temperature trends for the two samples or regions are behaving the same, and that any data errors in the temperature datasets (which are likely to be numerous) are not significant and so are not in need of correction or adjustment.

In Post 57 I used this approach to compare the temperature trends of neighbouring countries in central Europe (Germany, Czechoslovakia, Austria and Hungary). The results showed that if the MTA for a country was determined using data from more than about fifteen different station records then there was little difference between MTAs for different countries, and thus very little error in the MTA of each country. This is because of a property of statistics called regression towards the mean. This basically states that if any dataset contains errors in its measurements (which most data does), and those errors are random in their size and distribution (which they often are), then the errors will tend to cancel each other when you average the data. Moreover, the more data you average, the greater the cancellation of errors and so the more accurate will be the result. If errors don't cancel, then that is because the errors are systematic not random, so the process also helps to identify these as well.

In Post 67 I repeated this process for temperature data from the USA. In this case instead of comparing data from adjacent regions I compared different samples of one hundred stations from the same region: the entire contiguous United States. The result was the same as in Post 57 with each sample exhibiting an identical temperature trend over time with identical fluctuations in the 5-year moving average of the trend.

In this post I will repeat the country comparison of Post 57 but using the 5-year moving average of the temperature trend data from the four neighbouring Scandinavian countries of Norway, Sweden, Finland and Denmark. These trends were determined in Post 135, Post 136, Post 137 and Post 48 respectively. The results are shown in Fig. 139.1 below.

Fig. 138.1: A comparison of the 5-year average temperature trends since 1700 for Norway, Finland and Denmark compared to that of Sweden. The trends for Finland and Norway are offset by ±3°C for clarity.

In Fig. 139.1 I have compared the trends of Norway, Finland and Denmark with that of Sweden. The reasons for choosing Sweden as the comparator were both geographic and practical. It sits between the other three countries and so is a near neighbour for each (Finland and Denmark are not near neighbours so would not be good comparators). But it also has the most stations of the four countries and so should have the most reliable trend.

The data in Fig. 139.1 clearly shows that the trends for all four countries are very similar after 1900 but diverge as one looks further back in time towards 1800. The reason for this is the reduction in station numbers seen in each country as one moves back in time from 1950 (see Fig. 138.2 below). Given that it seems that somewhere between ten and thirty stations are needed in the MTA average in order for the errors to be minimized, we can see from Fig. 138.2 that this condition is satisfied for all four countries after 1890. That is why the MTAs diverge before 1890 but are very similar after that date.

Fig. 138.2: The number of station records included each month in the averaging for the mean temperature trends in Fig. 138.1.

If we just consider the data after 1850 we see that the agreement between trends for the different countries is remarkably good after 1890 (see Fig. 138.3 below). The agreement between Norway and Sweden, and Finland and Sweden are both particularly good to the point of their three trends being almost identical. There is also excellent agreement between Denmark's trend and that of Sweden after 1980 but less so before. This is probably the result of Denmark not only having much fewer stations than the other three countries, but also having fewer than ten stations before 1975.

Fig. 138.3: A comparison of the 5-year average temperature trends since 1850 for Norway, Finland and Denmark compared to that of Sweden. The trends for Finland and Norway are offset by ±3°C for clarity.

Summary

The data in Fig. 138.3 once again demonstrates the futility of temperature adjustments. The fact that the mean temperature anomalies (MTAs) of Norway, Sweden and Finland agree so well for over 120 years from 1890 onwards without data adjustments indicates that the averaging process alone can eliminate most errors.

The Denmark data also adds weight to the conjecture that between ten and thirty stations are needed in the average in order to eliminate most of the data errors. As the error size decreases with the square root of the sample size, an average of 25 datasets should decrease the error size by 80% (reducing each error to a fifth of its nominal value). 

Comparing the data of these four countries in this way also gives us more confidence in the determining the true nature of the regional temperature trend. All the data after 1900 pretty much agree so we can conclude that temperatures from 1900 to 1980 rose marginally by less than 0.3°C and then jumped by about 1°C in the 1980s. But this jump is still only comparable to the size of the fluctuations in the 5-year average.

From 1850 to 1900 both Denmark and Norway diverge from Sweden slightly but in different directions. But this is based on a comparison of only one or two stations in each case and so is not unexpected.

119: Greenland - temperature trends COOLING before 1990

Climate catastrophe is generally thought of as occurring in two ways: rising temperatures and rising sea levels. Greenland is fairly unique in that it is claimed by climate science to be indicative of both. The only problem is that in reality things aren't that simple. In fact Greenland is currently colder than it was 100 years ago.

The significance of Greenland is two-fold. Firstly it is the largest landmass in the Arctic Circle. As such it is one of the best indicators of climate change near the North Pole given that, as I showed in the previous post, there is no reliable temperature data within 840 km of the North Pole. But secondly, Greenland, like Antarctica, is a large store of frozen fresh water. Its ice sheet is second only in size to that of Antarctica, and has an average thickness of 1,500 m, rising to over 3,700 m above sea level at some points. If it were to melt completely it would raise global sea levels by more than seven metres. Yet between 1930 and 1990 the climate of Greenland actually cooled by almost 2°C, and while the mean temperature has risen quite sharply by a similar amount since 1990, mean temperatures are still below their 1930 levels (see Fig. 119.1 below).

Fig. 119.1: The mean temperature change for Greenland since 1920 relative to the 1976-2005 monthly averages. The best fit is applied to the monthly mean data from 1931 to 1990 and has a negative gradient of -2.81 ± 0.35 °C per century.

In order to quantify the changes to the climate of Greenland the temperature anomalies for each of the 39 stations with the most data (over 300 months) were determined and averaged. This was done using the usual method as outlined in Post 47 and involved first calculating the temperature anomaly each month for each station, and then averaging those anomalies to determine the mean temperature anomaly (MTA) for the region. The MTA since 1920 is shown as a time series in Fig. 119.1 above and clearly shows that temperatures declined continuously from 1930 to 1990 before rebounding.

The process of determining the MTA in Fig. 119.1 involved first determining the monthly reference temperatures (MRTs) for each station using a set reference period, in this case from 1976 to 2005, and then subtracting the MRTs from the raw temperature data to deliver the anomalies. If a station had at least twelve valid temperatures per month within the MRT interval (1976-2005) then its anomalies were included in the calculation of the mean temperature anomaly (MTA). The total number of stations included in the MTA in Fig. 119.1 each month is indicated in Fig. 119.2 below. The peak in the frequency after 1980 suggests that the 1976-2005 interval was indeed the most appropriate to use for the MRTs.

Fig. 119.2: The number of station records included each month in the mean temperature anomaly (MTA) trend for Greenland in Fig. 119.1.

The data in Fig. 119.2 above indicates that after 1960 there were up to 39 active stations, but before 1890 there were less than about five. As five is generally too low a number to produce a reliable trend, particularly over a large region like Greenland, the MTA data in Fig. 119.1 was truncated with only data post-1920 being shown. However, if all the data is considered, the MTA trend will have data extending back to 1863 as shown in Fig. 119.3 below. Note also that the low number of stations before 1940 results in a much higher variance of points in Fig. 119.3 about the mean (yellow line). This is more evidence of the greater unreliability of this earlier data, which is why the plot shown in Fig. 119.1 is more statistically reliable.

Fig. 119.3: The mean temperature change for Greenland since 1860 relative to the 1976-2005 monthly averages. The best fit is applied to the monthly mean data from 1871 to 2010 and has a positive gradient of +0.93 ± 0.12 °C per century.

The locations of the 39 stations used to determine the MTA in Fig. 119.3 are shown in the map in Fig. 119.4 below. This appears to show that the stations are all located on the coast, with none in the interior or at altitude, and a majority on the south-west coast. Of these 39 stations, five are long stations with over 1200 months of data before 2014, and a further eighteen are medium stations with over 480 months of data. What is more remarkable is how many stations Greenland has despite its low population. This may be because it has historically been part of the Kingdom of Denmark. The result is it has a similar amount of temperature data as Denmark (see Fig. 48.4 and Fig. 48.5 in Post 48) yet its population is only 1% of that of Denmark.

Fig. 119.4: The (approximate) locations of the 32 longest weather station records in Greenland. Those stations with a high warming trend between 1911 and 2010 are marked in red while those with a cooling or stable trend are marked in blue. Those denoted with squares are long stations with over 1200 months of data, while diamonds denote stations with more than 300 months of data.

If we next consider the change in temperature based on Berkeley Earth (BE) adjusted data we get the MTA data shown in Fig. 119.5 below. This again was determined by averaging each monthly anomaly from the 39 longest stations in Greenland. The mean temperature follows a similar trajectory to that of the unadjusted data in Fig. 119.3 with temperatures fluctuating by over 1°C and a large peak occurring around 1930. However the BE adjustments appear to have lowered this peak relative to temperatures in 2010 by over 0.5°C compared to the raw data in Fig. 119.3.

Fig. 119.5: Temperature trends for Greenland based on Berkeley Earth adjusted data. The best fit linear trend line (in red) is for the period 1876-2010 and has a positive gradient of +1.43 ± 0.07°C/century.

Comparing the curves in Fig. 119.5 with the published Berkeley Earth (BE) version for Greenland in Fig. 119.6 below shows that there is good agreement between the two sets of data. This indicates that the simple averaging of anomalies used to generate the BE MTA in Fig. 119.5 is as effective and accurate as the more complex gridding method used by Berkeley Earth in Fig. 119.6. In which case simple averaging should be just as effective and accurate in generating the MTA using raw unadjusted data in Fig. 119.1 even though the geographical distribution of stations is far from homogeneous, as was shown in Fig. 119.4.

Fig. 119.6: The temperature trend for Greenland since 1820 according to Berkeley Earth.

Most of the differences between the MTA in Fig. 119.3 and the BE versions using adjusted data in Fig. 119.6  are mainly due to the data processing procedures used by Berkeley Earth. These include homogenization, gridding, Kriging and most significantly breakpoint adjustments. These lead to changes to the original temperature data, the magnitude of these adjustments being the difference in the MTA values seen in Fig. 119.3 and Fig. 119.5. 

The magnitudes of these adjustments are shown graphically in Fig. 119.7 below. The blue curve is the difference in MTA values between adjusted (Fig. 119.5) and unadjusted data (Fig. 119.1), while the orange curve is the contribution to those adjustments arising solely from breakpoint adjustments. Both are considerable after 1920 with the former leading to an additional warming since 1930 of up to 0.5°C. These adjustments are, however, much smaller in total than the natural variation of 2°C seen in the raw data in Fig. 119.3, so while they change the overall magnitude of the climate changes slightly, the general form of the temperature trends in Fig. 119.5 and Fig. 119.3 look broadly similar.

Fig. 119.7: The contribution of Berkeley Earth (BE) adjustments to the anomaly data in Fig. 119.5 after smoothing with a 12-month moving average. The blue curve represents the total BE adjustments including those from homogenization. The linear best fit (red line) to these adjustments for the period 1921-2000 has a positive gradient of +0.56 ± 0.02 °C per century. The orange curve shows the contribution just from breakpoint adjustments.

Summary

According to the raw unadjusted temperature data, the climate of Greenland has cooled from 1930 to 1990 by about 2°C. It then warmed by a similar but slightly smaller amount until 2005 (see Fig. 119.1).

Over the same period adjusted temperature data from Berkeley Earth appears to show that the climate of Greenland has warmed by over 0.5°C since 1930 and up to 3.5°C since the 1880s (see Fig. 119.5).

The reliability of the temperature data before 1930 is debatable due to the low number of stations and the large jumps in temperature that occur repeatedly. The origin of these jumps is uncertain but cannot solely be the result of greenhouse gas emissions when those emissions increased the atmospheric carbon dioxide concentration by so little compared to today. However, similar patterns are seen in the temperature data of nearby islands of Iceland and Jan Mayen (from 1920 only), so these features seen in the data before 1930 may be real changes to the climate and not localized data errors.

Acronyms

BE = Berkeley Earth.

MRT = monthly reference temperature (see Post 47).

MTA = mean temperature anomaly.

Link to list of all stations in Greenland and their raw data files.

48. Denmark - temperature trends STRONG WARMING 1.8°C

In total, Denmark has twenty-two sets of temperature data that exceed 480 months in length (see here). Of these, eight contain over 1200 months of data (long stations), with the longest being Copenhagen (Berkeley Earth ID: 154574) which has continuous data from 1798, and some data fragments that go as far back as 1768. This suggests that the country has a similar number of station temperature records as New Zealand (see Post 8), but surprisingly it is less than is found for the Danish autonomous territory of Greenland which has a population of less than 60,000 and which I will look at in detail at some point in the future.

Fig. 48.1: The locations of long stations (large squares) and medium stations (small diamonds) in Denmark. Those stations with a high warming trend are marked in red.

The distribution of these weather stations in Denmark is indicated in Fig. 48.1 above. It shows a fairly even spread that covers most of the country. It also shows that most of the station data exhibits some significant degree of warming, with only two stations exhibiting a cooling trend (defined as being a trend that is less than twice the uncertainty in the trend).

The data from Denmark is interesting in one other respect in that, of its fourteen medium length station temperature records (i.e. those with over 480 months of data but less than 1200), four have no data after 1970 but do have data going back to the 19th century, while six have no data before 1970. This means that these two groups of stations require different time periods for the calculation of the reference temperatures needed to find their monthly temperature anomalies. For an explanation of the rationale and process used to determine the temperature anomalies via the calculation of monthly reference temperatures (MRTs), please refer to my previous post.

 

Fig. 48.2: The maximum amount of temperature data available from Denmark each month for inclusion in the mean temperature trend.

This problem is illustrated in Fig. 48.2 above. The two peaks in the frequency distribution indicate the two different possibilities for the MRT period. As I pointed out in Post 47, ideally the MRT period needs to be about 30 years in length with at least 40% data coverage. One way to circumvent this problem is to calculate the temperature trend for two MRT time intervals (the data in Fig. 48.2 suggests that 1891-1920 and 1971-2000 should be optimal), compare the results, and if necessary take a weighted average. 

 

Fig. 48.3: The temperature trend for Denmark since 1750. The best fit is applied to the interval 1851-2000 and has a positive gradient of +1.18 ± 0.09 °C per century. The monthly temperature changes are defined relative to the 1971-2000 monthly averages.

If we choose 1971-2000 as our reference period for the MRTs, then the overall temperature trend is as shown in Fig. 48.3 above. The number of stations included each month in this overall trend is shown in Fig. 48.4 below. Overall, up to seventeen stations are included, but before 1970 that drops to less than ten with only one station with data before 1870. The result is that there appears to be a fairly continuous warming trend from 1851 to 2000 as indicated by the data in Fig. 48.3.

Fig. 48.4: The number of sets of station data included each month in the temperature trend for Denmark when the MRTs are calculated for the period 1971-2000.

Now consider what happens if we choose 1891-1920 as our reference period for the MRTs. The result is that there are more stations included before 1970, but less after (see Fig. 48.5 below). This also changes the form of the temperature trend in Fig. 48.6.

Fig. 48.5: The number of sets of station data included each month in the temperature trend for Denmark when the MRTs are calculated for the period 1891-1920.

What we now see in Fig. 48.6 is a much smaller warming trend before 1980 (less than 0.6 °C with possibly higher temperatures before 1800), but a more pronounced jump in temperatures after 1988. This similar to the trend seen for South Africa (see Post 37) and also for Europe as a whole (see Post 44). It is important to note, though, that all the data before 1860 in both Fig. 48.6 and Fig. 48.3 comes from just one station record: Copenhagen (Berkeley Earth ID: 154574). This means the accuracy and reliability of this data cannot be truly ascertained.

Fig. 48.6: The temperature trend for Denmark since 1750. The best fit is applied to the interval 1768-1980 and has a positive gradient of +0.30 ± 0.05 °C per century. The monthly temperature changes are defined relative to the 1891-1920 monthly averages.

The analysis outlined above means that we have two possible results for the temperature trend in Denmark. Both are fairly similar, and for once both are in general agreement with the trend published by Berkeley Earth (see Fig. 48.7 below). But can we combine them into a single result?

Fig. 48.7: The temperature trend for Denmark since 1750 according to Berkeley Earth.

The answer is yes. If we take the weighted average of each of the two trends in Fig. 48.3 and Fig. 48.6 we get the result shown in Fig. 48.8 below. The relative weightings of each month's data is determined by the number of stations included in the average for that month as indicated in Fig. 48.4 and Fig. 48.5 respectively. There is, though, one other factor we need to take into account: the different MRT intervals for the two original trends. Without a correction term this will distort the final data.

In order to allow for the differing MRTs, the trend curve in Fig. 48.3 needs to be first adjusted upwards so that the mean temperature anomaly for the period 1891-1920 is zero in order to be consistent with the data in Fig. 48.6. This requires an upward adjustment of 0.634 °C. Only after this adjustment has been made can the weighted average be determined.

Fig. 48.7: The weighted temperature trend for Denmark since 1750. The best fit is applied to the interval 1851-2000 and has a positive gradient of +1.02 ± 0.09 °C per century. The monthly temperature changes are defined relative to the 1891-1920 monthly averages.

Conclusions

The data from Denmark appears to show a warming trend of about 1.5 °C since 1850. This is by far the largest warming seen in any of the regional records that I have investigated so far. It is also one of the few that appears to agree with IPCC reported values. However, this is not as straightforward as it seems. For a start changing the MRT interval from 1971-2000 (as in Fig. 48.3) to 1891-1920 (as in Fig. 48.6) dramatically reduces the temperature trend before 1980. So which one is correct? 

Then there is the problem of the sudden jump in temperature in 1988 of 0.93 °C. A similar jump was seen in the temperature trend for the Europe data in Post 44 as well. The reason for this is still unclear (at least to me). In Post 45 I speculated that it could be the result of improved air quality in Europe due to EU legislation. Alternatively, it could be the consequence of a change in measurement method, such as a change from liquid-in-glass thermometers to electronic systems which occurred around that time. What is strange is the timing and suddenness of this increase. 

Whatever the true scale of the temperature rise in Denmark since 1750, it cannot be explained entirely by direct anthropogenic surface heating (DASH) or waste heat. The best estimate of the expected magnitude of DASH for Denmark (based on its population density) is about 0.35 °C since 1850. However, this could be greater if the source of the heating from human industrial activity is concentrated around the locations of the major weather stations. For example, a city like Greater London with an area 1569 km² consumes over 132 TWh of energy each year. This equates to a power density of 9.6 W/m², or an effective temperature rise of over 4 °C. Clearly something similar but less extreme could be occurring around the major cities in Denmark, but at the moment, without direct measurement data, that remains as speculation.