I have been studying differences in climate data between those  areas  of the world with very low atmospheric water vapour (Deserts and Polar regions) and those areas with very high water vapour content (Tropical Wet regions).  The data set consists of all 5500 station data corresponding to CRUTEM4 kindly provided by the UK Met office. Each station has then been classified by climatology using it’s geographic location and a lat,lon grid based on the Köppen-Geiger climate classification [1].

We define ARID stations as all those situated in Deserts or Polar regions i.e. those in areas with precipitation ‘W’ or climate ‘E’ in [1]. These areas have the lowest atmospheric water content on Earth. The WET stations instead are defined as those within Tropical fully humid areas – ‘Af” in [1]. These include tropical rain-forests and  have the highest atmospheric water vapour content on Earth.  In a previous post I already showed that the temperature anomalies in the Sahara rose faster than a similar area in S. Asia, so I wanted to extend this study globally using the latest CRUTEM4 data.   I therefore calculated the global anomalies for both sets of stations ARID and WET using the same algorithm as used for CRUTEM4. The results are shown in Figure 1.

Fig 1: Temperature anomalies for ARID stations in red and WET stations in blue. The smooth curves are FFT smoothed curves. The black dashed curve is an FFT smooth to the full CRUTEM4 temperature anomalies.

There is a clear trend in the data that ARID stations cool faster and warm faster than WET stations. They seemingly react stronger to external forcing. The WET humid stations respond less than  both the ARID stations and the global average.  The location of the stations are shown in Figure 2 which is taken from reference [1]. The ARID stations are the yellow desert areas and the light blue polar areas. The WET stations are located in the  red zones – Amazon, central Africa and SE. Asia.

Fig 2:Climatic zones defined by KÖPPEN-GEIGER classification. DRY in yellow+polar regions. WET in red. see: http://koeppen-geiger.vu-wien.ac.at/

We will assume that there are external forcings on the climate both anthropogenic and natural which are reflected in temperature anomalies. If we label the forcing DS and the consequent  change in temperature anomaly as DT.

DT1 = g1*DS   for  ARID   and   DT2 = g2*DS    for  WET. We then divide the last 110 years of data into 3 main time  periods to measure DT1 and DT2, and assume that DS is a universal global external forcing ( for example due to CO2 )

Critics may argue that heat inertia effects due to nearby oceans are causing tropical climates react slower than desert regions. However, the IPCC argues that  feedbacks from increased water evaporation will lead to enhanced warming. This is not observed in those regions most affected by water vapour. In fact the opposite seems to be the case implying negative feedback. If we make the assumption that there have been 3 separate forcings for the 3 time periods above and that there is no other difference other than humidity, then we can estimate the water feedback F.  Taking F=0 for ARID stations:

(DT2-DT1) = F*DS  ;    where DT1 = g0*DS   and DT2 = (g0+F)*DS

F/g0 for the 3 periods :  -0.5+-0.1   ,  -0.7+-0.1  ,  -0.3+-0.1

Average Feedback parameter  =   -0.5 (+- 0.1)*g0  which taking G0 =  the Stefan-Boltzman value 4*sigma*T^3 = 3.75W/m2K-1.

Water Feedback =  - 1.8 +- 0.2 W/m2K-1  

Remarkably this is the same value as that derived from  a simple argument regarding the Faint Sun paradox see here. It has been pointed out by Richard Lindzen [2]  that much of the Earth’s heat is transported bodily through evaporation and convection to the upper atmosphere where IR opacity is low and  can then escape to space. Therefore water feedback effects depend on the water vapor content of the upper atmosphere more than that at the surface.  Increased evaporation, convection and rain out could even dry out the upper atmosphere. This could be a possible mechanism for negative feedbacks.  Such effects would be much smaller in ARID areas with little or no evaporation.

1. Rubel, F., and M. Kottek, 2010: Observed and projected climate shifts 1901-2100 depicted by world maps of the Köppen-Geiger climate classification. Meteorol. Z., 19, 135-141. DOI: 10.1127/0941-2948/2010/0430. http://koeppen-geiger.vu-wien.ac.at/

2. Richard Lindzen, Some uncertainties with respect to water vapor’s role in climate sensitivity. Proceedings NASA workshop on the role of Water Vapor in Climate Processes.

New Stations Kyrgyzstan

How 2005 & 201o got “ranked” the warmest years.

The new temperature data from  CRUTEM4 has added 628 new weather stations, including strangely enough over 50 from Kyrgyzstan. Most  of these stations are in far northern latitudes. There are none in the Southern Hemisphere. The general perception is  that the greenhouse effect is a global phenomenon and  all parts of the world will experience  ”climate change”. However GISS, UAH and Hadcrut trends show recent larger increases in the Arctic.  Antarctica on the other hand shows little sign of any warming. The Arctic is surrounded by land masses and this is precisely where the vast majority of  the new station data have been added by CRU. CRUTEM4 has significantly increased sampling across the (still very cold) Arctic borders. The best way to see this is through the effect it has on global average temperatures. I will ignore  all the counter-arguments as to why  ”global” temperatures cannot be shown. These are the temperatures to which  the station anomaly measurements refer and  reflect their geographic distribution. Figure 1 shows the area averaged temperatures for CRUTEM3 compared to CRUTEM4.

Fig 1: CRUTEM4 and CRUTEM3 annual average temperatures. Note how including the arctic stations has reduced the average temperatures in the northern hemisphere. Note also temperature spike in 2009.

The effect of adding so many Arctic stations is to essentially  drop the average temperatures in the northern hemisphere by up to 1-2 degrees C. This is accentuated during  recent years where more of the stations have data. What effect does all this have on the temperature anomalies ? The anomalies for CRUTEM4 will naturally tend to increase slightly over CRUTEM3 for two reasons.

1. The addition of stations in an area which is already known to show  strong warming will lead to  a higher global average anomaly as  available grid points get filled in preferentially there.

Concentration of grid points and land masses close to the North Pole

2. The density of grid points in a 5×5 degree grid increase rapidly as we get closer to the poles. In fact if you stand near  the north pole and simply walk around it you will then pass through 75 grid points. This means that there are far more grid points available in which to place new stations than for example at the equator. Furthermore because there is so much land area close to the north pole there are stations nearby unlike at the south pole. It is true that the area averaging does weight according to  latitude. However CRUTEM4 has filled as many grid points as possible just near the north pole. It is clear that this is the explanation as to why  the post 1998 anomalies have increased sufficiently for 2010 to become the “warmest year”, although within errors this is anyway meaningless. A detailed comparison of the anomalies after 1994  is shown in Figure 2.

Fig 2: Detailed comparison of temperature anomaly results from CRUTEM4 and CRUTEM3.

To look further into this  I compared missing grid points between CRUTEM3 and CRUTEM4. A missing grid point is simply one single 5×5 cell which does not contain any stations. There are 2592 cells in a 5×5 degree world grid of which perhaps 65% contain just ocean. If all available land points were covered with stations then there would be an ideal minimum of  ~1700 missing cells. Figure 3 shows the actual number of missing cells per year (and month) for CRUTEM3 and CRUTEM4.

Figure 3: Missing cells versus year for CRUTEM3 and 4. Note spike in 2009.

The geospatial changes can be seen in more detail by comparing the ratio of sampled cells from the tropics (LAT < 25) with those at large latitudes (LAT >25) – see Figure 3.

Fig3: Ratio of sampled cells from the Tropics (LAT<25) and higher latitudes

CRUTEM4 accentuates further the sampling bias away from the Tropics.  It is pretty clear that  oversampling of  the arctic region  leads to an  increase in the “global” temperature anomaly, as is now “measured” with CRUTEM4.  However,  it should be remembered when reading various press releases and news headlines that  the error on  a single annual anomaly value is ~ 0.1 deg.C , so statistically it is meaningless to state that 2010 is warmer than 1998 or vice versa.

P.S. I wonder what is the real origin of  the spike in 2009?  Why did  so many of the new arctic stations suddenly  disappear that year ?

UPDATE 23/3/12: I have discovered that 110 stations have changed numbering between CRUTEM3 and CRUTEM4 ! Therefore the map below has been revised to show just the new and discarded stations excluding those renumbered. The list of renumbered stations can be found here.

The new data from the Hadley Centre and UEA CRU  (HadCrut4) has been released.  The temperature anomalies calculated just from the station data is called CRUTEM4. I have downloaded all the new station data and compared the results with the previous iteration CRUTEM3.  What has changed?

The most obvious difference is that many new stations have been added , while many others have been dropped. There are now 5549 stations in the set compared to 5097 in CRUTEM3. 738 628 new stations have been added while 286176 stations have been discarded.  We can see exactly which stations have been added and lost in the map below. The red points are the new stations added in CRUTEM4, while the blue points are those discarded from CRUTEM3. Drag a rectangle in the map to zoom in. Click on a station to view data.

Clearly the new points are all in the northern Arctic regions especially over Russia. Some of the  dropped stations have few updates since 2000, but whether that means they no longer exist is unclear. Many stations have been dropped from Northern America.

What is the effect of this on temperature anomalies ? I calculated the anomalies for both data sets using the Hadley provided scripts and compare both results below.

Fig 1: CRUTEM4 and CRUTEM3(blue) temperature anomalies

There is no real statistical difference beween the two results. Small differences that are apparent are concentrated at end of the 19th century (slightly lower) and more contentiously (slightly higher)  during the last 10 years. HADCRUT3 has shown no warming since 1998 and this has become something of an embarrassment to the AGW narative. A zoom in to the last 10 years identifies the subtle new change.

Fig2: Closeup of the differences between CRUTEM4 and CRUTEM3(blue)

CRUTEM4 now places  2010 a tiny bit warmer than 1998, but there is no statistical significance to this despite the propaganda value. Note also that this result covers  just the land station data. HADCRUT4 also includes sea surface temperature data form the Hadley Centre which have also been updated. This may then explain other differences as discussed for example here.  For details of how the sea surface data has been updated – see the Met Office Web site.

1. Tackling Climate Change
2. Reducing Emissions
3. Meeting Energy Demand

I would have placed these priorities in exactly the reverse order since cheap reliable energy is the key to prosperity while climate change is of little direct importance for the UK economy. Meanwhile DECC is spending eye watering amounts of money “tackling climate change” as sumarised in their “Cost Benefit analysis” from 2008 (see fig 1).

Fig 1: CLIMATE CHANGE ACT 2008 IMPACT ASSESSMENT

Spending this sort of money one would expect large long term benefits. The assumption seems to be that through example the UK will convince the world to abandon growth and cut carbon emissions to avoid planetary disaster. As I see it there are two basic problems with this noble position.

1. For the UK itself there has been no  discernable  change in  climate since 1940 and it is unlikely that significant change will occur by 2050 either. The evidence for this comes from the HADCRUT3 averaged temperature anomaly data for  all UK stations from 1940-2011. The result is shown in Figure 2. There has been essentially no change.

Figure 2: UK average temperature anomaly 1940-2011. The blue curve and the red points show the annual variation. The dashed curve shows the monthly variations.

1.  In 2006 China increased CO2 emissions over 2005 levels by 545.2 Mt, while in the same year total UK emissions were just 535.8Mt. The 2008 climate change act aims to cut UK emissions to 20% of 1990 levels by 2050 at a costs of hundreds of billions of pounds. It will have no effect on UK temperatures and globally its effect is also insignificant. The UK’s contribution over 40 years will be to offset just 1 year of increases in China’s CO2 emissions.

If on the other hand the aim is to curb UK dependence on fossil fuel imports, then it would be far better to invest just a fraction of this money into nuclear fusion. Only high power densities can replace fossil fuels, and the only non-carbon alternatives are nuclear fission and nuclear fusion. Renewables like wind and solar will always have far too low power densities to be of any significance.

There are indeed many large cities in the list which have grown over the last century, particularly Sao Paolo and Beijing.So at first glance this looks like direct evidence that urban warming could well be skewing the overall result. To investigate this possibility I excluded all the cities in the above list with a population greater than 0.4 million and recalculated the global average anomalies over the full time span. The result is surprising – there is no effect – see figure 1! It is indeed true that urban warming does not systematically effect the overall trend in global temperatures.

Figure 1: Comparison of the land station data CRUTEM3 with and without the warmest cities listed in the table.

So the conclusion is more or less the same as the BEST result. Observed increases in global temperature (anomalies) are not  effected by large urban areas.  I am convinced.

Colour    Anomaly change(Deg.C)
Red           AA > 1.0
Orange        0.4 < AA < 1.0
Yellow        0.0 < AA < 0.4
Cyan          -0.4 < AA < 0.0
Blue          -0.4 < AA

It immediately becomes obvious  that  the bulk of  observed warming is concentrated in the Northern Hemisphere : Eastern Europe, Russia, central Asia, India, China, Japan, Middle East, North Africa. These are all areas of rapid population increase, development and industrialisation. There is essentially no warming at all in the Southern Hemisphere. Bolivia, Peru, Paraguay and Argentina all appear to be cooling. Even Australia and Zealand are static or  cooling. The US is evenly divided and the UK shows essentially no signal at all.

Fig 1: Locations of weather stations

The US actually has hundreds of stations used for the anomaly calculations – so most of them show little (<0.4deg.C) or no warming since before 1960 – see figure 1.

Could much of the observed temperature rise over  the last 6 decades  be simply due to increasing urbanisation and development since ~1960 ? We know that many of the weather stations are close to urban areas, so let’s do a back of the envelope calculation to see if this is a realistic possibility.

Estimates of the total urban land cover globally from satellites are about 2% of total land surfaces – approximately 3,000,000 km2 [1]. Total average world energy consumption rate ( fossil, nuclear,hydro etc.) is about  15 TW ( see: wikipedia), and increasing by ~ 5%/year. My guess is that 80% of this energy ends up as heat (2nd law thermodynamics). Assuming that  energy consumption is concentrated mostly in urban areas then the net “anthropogenic”  heating in those areas works out at around 5 watts/m2. This  then leads to  an average 1.4 degreeC. rise in temperature for urban areas.  Anyone who has lived in the city knows from experience that the surrounding countryside is indeed  some 1-2 degrees colder.

This expansion in urbanisation is accelerating and by 2030 global urban land cover will increase by between “430,000 km2 and 12,568,000 km2, with an estimate of 1,527,000 km2 more likely”.[quoted from ref: 1]. The main growth areas are in China, India, South West Asia and Africa, which is also where the station data show most warming. Those stations appear to be experiencing anthropogenic warming of a rather more direct kind than that proposed by IPCC. They are likely warming due to increasing urban heating.

Another result of the debate on WUWT was the conclusion that there are actually 2 different ways to determine global temperatures. The first is to simply average the measured temperatures within a lat,lon grid and then weight the result according to the surface area  on the Earth. The second method is to average instead the black body equivalent energy flux (T^4) to derive an equivalent “radiative” temperature. The averaged T^4 term is then converted back to a single “radiative” temperature by taking the 4th root. This would be interpreted as the black body temperature of the Earth. This new second method however has the effect of biasing the contribution of places with higher than average temperatures.  I used both methods to calculate global temperatures from 1950 to 2011. A comparison of the two results  for the  period 1950-2011 is shown below.

Fig 2: Average temperatures for Southern Hemisphere(SH) and Northern Hemisphere(NH) . The red curves are based on average (4Root((t+273)^4)-273. The black curve is the area average of measured temperatures. Note how 1) SH temperatures remain stable. 2) Just NH warms and T^4 average is ~0.1 deg.C higher.

The conclusions from this study are:

1)  There has been no warming in the Southern Hemisphere since 1950. Radiative and measured temperature averages also agree.

2) Warming is observed in the Northern Hemisphere since 1950. This is concentrated in regions where rapid development and urbanisation are also occurring. It is therefore probable that some of these “hot stations” are  affected by urbanisation warming rather than AGW.  Discrepancies between the radiative and measured temperatures also imply that a few main Hot Spots are responsible.

Acknowledgements:  1) Station data are from UK Met office     2) Flash map interface is by DIY-Map

Background: In 200o an IPCC special report proposed several future economic scenarios each with a different CO2 emission profile. For the 2007 assessment report these scenarios were used to model predictions for future global temperatures. The results for each of the scenarios were then used to lobby governments. It would appear that as a result of these predictions, there is  one favoured scenario – namely B1 which alone is capable of limiting temperature rises to 2 degrees.

The Scenarios: These descriptions are taken from the SRES special report.

“The A1 scenario is a case of rapid and successful economic development, in which regional average income per capita converge – current distinctions between “poor” and “rich” countries eventually dissolve. The transition to economic convergence results from advances in transport and communication technology, shifts in national policies on immigration and education, and international cooperation in the development of national and international institutions that enhance productivity growth and technology diffusion.

The A2 world has less international cooperation than the A1 or B1 worlds. People, ideas, and capital are less mobile so that technology diffuses more slowly than in the other scenario families. International disparities in productivity, and hence income per capita, are largely maintained or increased in absolute terms. People, ideas, and capital are less mobile so that technology diffuses more slowly than in the other scenario families. International disparities in productivity, and hence income per capita, are largely maintained or increased in absolute terms.

The central elements of the B1 future are a high level of environmental and social consciousness combined with a globally coherent approach to a more sustainable development (favoured by IPCC?).  Heightened environmental consciousness might be brought about by clear evidence that impacts of natural resource use, such as deforestation, soil depletion, over-fishing, and global and regional pollution, pose a serious threat to the continuation of human life on Earth. A strong welfare net prevents social exclusion on the basis of poverty.”

The consequent CO2 emission trends  which have been  simulated  for each scenario are shown  below.

Fig 1: CO2 levels for different scenarios

IPCC approved models were run on these scenarios using these predicted CO2 levels. As discussed before all IPCC models assume a strong positive feedback of water leading to amplifications of 100-200%. The resultant predictions over a 300 year period are shown below. These graphs undoubtedly helped influence political opinion to limit future warming to 2 degrees which implicitly supports scenario B1. Note also how scenario A2 explodes exponentially, presumably leading to the extinction of all life on Earth. This is the only scenario without some  world eco-governance body  and seemingly ends in disaster !

Figure 2: AR4 figure for long term predictions for each scenario

Basics: The main focus of this post is on  the technical summary of WG1 which contained specific short term  predictions using the same models  for warming up to 2030. This is important because good science makes testable predictions over realistic timescales. Otherwise it is not science at all but just dogma. The data used in  the original 2007 report was available only up to 2005. Since then we have had 6 more years of data which we can now confront with the model predictions. Meanwhile  CO2 levels have continued to rise in line with all scenarios (except that fixing levels at  2000).

The figure below shows the new data points plotted over the original figure that appeared in the WG1 report ( see here). The new black trend curve is a smoothed FFT fit through the data points. The results are startling.

Figure 3: TS figure from WP1 updated with the latest temperature data from HADCRUT3. The black curve is an FFT smooth through all points. Curves are (quoting TS): Multi-model means of surface warming (compared to the 1980–1999 base period) for the SRES scenarios A2 (red), A1B (green) and B1 (blue), shown as continuations of the 20th-century simulation. The latter two scenarios are continued beyond the year 2100 with forcing kept constant (committed climate change as it is defined in Box TS.9). An additional experiment, in which the forcing is kept at the year 2000 level is also shown (orange).

The trend speaks for itself. Predicted warming has not occurred and the actual temperatures are all more than one standard deviation below even the fixed 2000 CO2 levels (orange curve). All 6 annual temperatures lie below all scenario curves. The quoted error on a single measurement is 0.05 deg.C so we can now calculate the probability of these measurements  being a statistical fluctuation.

year sigma     probability
2006   1        0.32
2007   3        0.001
2008   4        0.0001
2009   2        0.04
2010   2        0.04
2011   6        <0.00001

The total probability that IPCC predictions are correct but the data points are just a fluctuation is vanishingly small ~ 10^-14 ! It is therefore possible to state with over 90% confidence that the IPCC 2007 model predictions are incorrect and exaggerate any warming. Will we have to wait another year for the 2012 data to be published before the IPCC admit that they have simply got it wrong ?

You can also download a poster about this post

AGW models predict “dangerous” global warming of 2-5 degrees this century by assuming net positive climate feedback values due to water of ~ 2watts/m2/degreeC [2]. Water feedback is  predicted to amplify an otherwise modest GHG rise of ~1 degree C  by some 200-500%. The main mechanism argued for such large water feedbacks is the increased evaporation with temperature leading to  enhanced H2O greenhouse effects.  Is there any direct evidence that such large positive feedbacks are already occuring ?

To look into this I selected weather stations[1] from 2 different regions: A) The Sahara desert with  very low humidity and B) South East Asia with high humidity. The assumption is that any H2O feedback will be present in B) but not in A). The two areas were chosen with exactly the same area and the same latitude:  [15-28, -12-30] and [15-28, 120-140] and only stations with records from before 1950.  The saharan stations were listed earlier here, and the 47 South East Asia stations are listed as a footnote [3].

The average of temperatures over both regions are shown in Figure 1. There are clear differences both in the annual extremes and in the variations essentially showing the effects in Asia caused by the presence of large sources of water vapour. The Sahara temperatures have large daily and annual swings in temperature because there is little greenhouse effects from water vapour.

Figure 1: Comparison of average temperatures across 17 Saharan stations and 42 "oceanic" Asian stations at the same latitude.

I have argued elsewhere that the way anomalies are calculated can effect the observed temperature rises over time. However for this study, we will use the standard CRUTEM anomalies which are based on the seasonal difference between measured values and monthly “normals” averaged over the period 1960-1991. Figure 1 shows how large the annual (and daily) variations can be in the Sahara(20 degree swings). Despite this we define the average temperature change as being the average of all individual station anomalies in each region. The result is shown in Figure 2:

Average Anomalies for each region. The monthly data has been smoothed by a fourier fit and then fitted with a linear dependence DT = m(x-1950) + CV. These fits are shown by the two lines through each set of points, and are independent of the smoothing process.

The averaged anomalies have been smoothed using a factor 5 FFT filter. This is only to reduce the point to point noise. The lines are linear fits through the smoothed data. In effect all we can say is that over the last 60 years in each case tempartures have risen by

S.E. Asia :   0.5+- 0.2  degrees C.   (error from fit)

Sahara:    1.0 +- 0.2 degrees C.      (error from fit)

These can now be used to derive an  effective water feed back value  assuming that it is active in Asia but absent in the Sahara.

Feedback for a change in radiative forcing DS

DS is assumed to represent AGW and any natural effects including feedbacks other than water vapour and  that it is the same in both regions.

For the Sahara   DS.G0 =  1.0 (+- 0.2)

For Asia              (DS+F).G0 = 0.5 (+- 0.2)

This yields    F = -0.5DS   (+- 0.3)

DS would need to be around 3.6 watts/m2 to cause a global rise in temperature of 1 degree. Therefore the final result for the feedback value of water with the above assumptions is

F(water) = -1.8 (+-1.1) watts/m2/degree C.

This result implies that the net climate feedback (sensitivity) to water is negative  or at most zero. The data is incompatible with large  positive values of +2.0 watts/m2/degreeC. as used in the majority of GCM models.  The fact that 2 billion years ago  temperatures were not much different than today despite 20% less solar radiation implies  that the Oceans stabilize climate through negative feedback (see here).

References
[1] Hadley Cru Station Data are available here
[2] S. Bony et al. How well do we understand and evaluate Climate Change Feedback Processes, Journal of Climate, Vol 19, P. 3445, 2006
[3] S.E. Asia stations used are listed here.

I have been studying the station temperature data used in the Hadley Cru global temperature analysis. These data consist of all the land based stations dating back to about 1700 containing monthly averaged temperatures for each year of data. These are the data used to produce CRUTEM3 and HadCruT3. Hadcrut3 includes sea surface temperatures. I have been using and extending the PERL analysis programs kindly provided by Hadley – available here. As a by-product, I also developed a  geographic station browser allowing you to view and plot all individual station – for further details see also here. A comparison of the calculated anomalies based on the >3000 stations and the usual HadCRUT3 is shown below.

Fig 1: The Red points are the temperature anomalies used by IPCC - Hadcrut3V. The black points are the calculation results of station-gridder.pl | make_globl_averages.pl calculated from all the station data (as described below).

There are small differences which confirm that the main component of the observed temperature rise is based on the land data. However it is important to understand exactly how these data points are derived.

Algorithm:
1. Average monthly temperatures are calculated at each station based on the following procedure: record the minimum and maximum temperature for each day and then calculate the average of the minimum and maximum. Calculate the averages for the month from these daily data. These are then recorded for each station file along with metadata [lat,lon] station name etc.
2. So-called monthly “normals” are calculated for each station by averaging each individual monthly temperature over the years 1961 to 1990. Standard deviations are also calculated. The normals are then assumed to represent a “standard annual variation”.
3. Anomalies are defined for each station by subtracting the monthly values for a particular month from these normal values. Stations without normals for 1961-1990 or where any anomaly is > 5 standard deviations are excluded.
4. The world is divided into a 5×5 degree grid of 1592 points. For each month the grid is populated by averaging the anomalies of any station present within each grid point. Most grid points are actually empty – especially for those early years. Furthermore the distribution of grid points with latitude is highly asymmetric with over 80 percent of all stations outside the tropics.
5. The monthly grid time series is then converted to an annual series by averaging the grid points over each 12 month period. The result is a grid series (36,72,160) ie. 160 years of data.
6. Finally the yearly global temperature anomalies are calculated by taking an area weighted average of all the populated grid points in each year. The formula for this is $Weight = cos( $Lat * PI/ 180 ) where $Lat is the value in degrees of the midle of each grid point. All empty grid points are excluded from this average.

Quality of the data

Fig 2: Distribution of stations: Left are stations with data dating back to 1850: Right are stations dating back to at least 1930

From 1850 -1860 less than 5% of grid points contain data (figure 3). This rises to 20% by 1940 and peaks at 30% from 1960 to 1990, before falling again to  23% currently. Figure 4 shows the latitude distribution which demonstrates that over 80% of stations lie outside the Tropics at high latitudes ( Europe, US, Russia, Australia etc.). This can also be seen visually in the map displays of the Flash application (Figure 2).

Fig3: The percentage of the 5x5 grid containing any station data

Fig4: Number of grid points at high latitudes > +-25 compared to grid points in the extended tropics < +- 25. Less than 20% of the sample are in the (extended) tropical region.

Observations:

1. There is poor coverage over the main tropical warm zone of the tropics +-25 degrees. The averages are biased to high latitudes with large summer to winter swings. This then also accentuates the temperature differences between the southern and northern hemisphere.
3. The stations are all on land and lack the sea surface temperature measurements. However the annual anomaly results are almost the same as those from Hadcrutem3VG which include sea surface temperature data. Therefore the land based temperature data dominate the temperature trends.

Experiments:

The first excercise I did was to look directly at the temperatures rather than at the anomalies. This also shows how an unbalanced averaging a high latitudes accentuates differences between North and South hemispheres and the annual variations. The monthly temperature data for the full period is shown in Figure 5.

Fig 5: Global Monthly temperatures. The global average = (SH+NH)/2 . The smoothed curves are 50 point running averages

You can see how initially the discrepancy between north and south diminishes as more tropical stations are added. They then separate again after about 1920 before narrowing again recently. Note also how it appears to be the reduction in the minimum temperatures (e.g. for January – North)   that seems to be the cause of the anomaly rise.

Could there still be systematic effects due to changes in  the distribution and proportions of stations over time ? Next I looked at the averaged temperatures for each of 3 regions A) -20 <lat <20 (Tropics) B) Lat >20 (Northern lats) C) LAT<-20 (Southern Lats). These results are shown below. The year 1863 actually had no measurements inside the tropics – hence the zero value.

Fig 6: Yearly averages for Tropics, Northern and Southern Lats.

There are a couple of observations here. Firstly note  how excluding the tropics changes the “global average temperature” (B+C)/2 by only about 1 degree.  Secondly, note how most of the  temperature rise since 1980 is concentrated in northern latitudes. I am assuming that the “experts” choose to work with anomalies rather than  absolute temperatures because of known systematic problems. Anomalies measure the deltas between monthly temperatures relative to a standard(normal) set. In effect we are subtracting two large numbers from each other and averaging the residues. It is assumed that if the Earth is warming overall by say 0.5 degrees , then the global average of all the deltas measured by each individual station will also rise by 0.5 degrees. We are no longer measuring the global temperature as such but rather changes of an evolving distribution of station measurements over several decades. The anomalies can still in principal be prone to systematic effects through over-sampling. A simple example of  how this could happen would be say if North America rose by 1 degree while simultaneously the Sahara fell by 1 degree-  maintaining no net increase in global temperatures. The averaging algorithm would then produce a net global increase in temperature because the US is over-sampled while the Sahara is under-sampled.

Normalisation Study

Next I looked at possible effects of the normalisation method used to extract temperature anomalies for each station. The normal procedure followed by HAD-CRU is to calculate monthly averages for each month and each station between 1961 and 1990. These are then used to calculated anomalies for each station, and average them in each grid point. I decided instead to use the actual temperatures at each grid point resulting from the average of these stations. Then in a second step I calculate the monthly normals for each month at each grid point by averaging all the available data. There is no particular reason  to take a fixed time period for the normals, since  anomalies are just deviations from the norm.  First I generated temperature grids from 1850 to 2010. I then used the grid monthly time series to derive normals per month for each grid point. Then we subtract the normal from  grid temperatures to derive anomaly grids. Finally the area weighted and annual averages are derived. How does this compare with the standard result ?

Fig7: The blue points are temperature anomalies from GRID point averages compared to the standard anomalies based on per station based averages from 1961-1990 (black points)

As you can see the time trend changes dramatically! The results show clearly that prior to about 1910 the temperature anomalies are highly dependent on the normalisation method. This is not surprising as the geographic coverage is  < 14%.  After 1920, however this analysis shows that we can have a high confidence level in the published values.

Conclusions

1. The coverage of stations is concentrated at relatively high(low) latitudes. There is far less coverage in tropical regions, which thereby tends to exaggerates seasonal global temperature changes.

2. Sampling problems over in  5×5 grid for early years must lead to systematic problems in early years because there are so  many  empty grid points.

3. The standard use of temperature anomalies per station over a fixed period 1961-1990 is  somewhat arbitrary. When anomalies are instead calculated against a longer reference period and use grid values rather than station values then significant diferences are seen prior to ~ 1920.

4.The  conclusion is that the land based data are reliable after ~1920 but that earlier data are subject to systematic errors. The absolute temperature data are also effected  when there is significant asymmetries in geographic sampling.