Mean annual temperature data for Luxembourg: 1951 - 2004
last update: 24 Aug 2005
This is a short ongoing report on the
trend in mean annual surface temperatures of Luxembourg, computed from the archive
GISS . Graphs and linear fits have been made including all years, and only years
1990 to 2004 with and without the heat-wave year 2003. The general annual trend
of the whole period is a temperature increase of about 0.0132 °C/year, what
gives 0.7 °C for the 54 year period. For the last 15 years (1990 - 2004) this
trend becomes 0.028 °C/year, an increase essentially caused by the 2003
heat-wave; omitting 2003, the trend is only 0.0106 °C/year. A comparison with
the Mauna Loa CO2 data gives a mixed picture for the relationship between local
temperature and global CO2.
It should be noted that the data come from Luxembourg's Findel meteorological station, located at the national airport in a region of ongoing urbanization and vastly increasing air traffic. The data have probably not been adjusted for an urban heat island effect.
1. Mean annual temperatures for 1951-2004, all data
2. Mean annual temperatures for 1951-2004, heat wave year removed
3. Mean annual trend for 1990-2004, all data
4. Mean annual trend for 1990-2004, heat-wave year 2003 left out
It comes as a surprise that the summer trend gets negative
for the 1990-2004 period, if the exceptional warm 2003 period is left out; the
winter trend is negative on that same period due to the cold 2003 winter, and
becomes positive if 2003 is omitted. This massive (albeit rather short-term)
heat wave event ups the mean decadal warming for 0.024 °C !
The annual trend of 0.0132 °C has to be compared to the global warming of about 0.0174 °C/year, as given by NasaGISS  for the (nearly) continuously warming period 1975-2004, or 0.0084 °C/year as derived from the satellite MSU measurements  from 1979 on. If the warming in Luxembourg would continue unchanged, the mean annual temperature increase for 2100 would be 1.98°C.
The surface temperature linear trend contrasts with the quasi exponential increase in global CO2 concentration, a fact often mentioned by many authors (for instance  ): this linear trend = 4.76*exponential growth factor (global CO2 concentrations see ).
Plotting surface temperature versus the monotonously increasing CO2 concentrations shows clearly two different periods during the last 45 years: from 1959 to 1987 mean temperatures decreased versus CO2 concentration, from 1988 to 2003 they increase:
This should be a clear warning in seeing an exclusive causal relationship between CO2 increase and warming: elementary physics tells us that CO2 is a greenhouse gas and should produce some warming; the data show that other factors beside CO2 must be acting! If we compute a climate sensitivity from the logarithmic fit to the 1988-2003 data, we find a temperature increase of 5.6°C for a doubling of the 1988 concentration (i.e. for 702 ppmV); this is purely speculative, as the period 1959 -1987 shows that the positive trend may revert to a negative one!
Hans Erren  gives the following formula to compute the warming dT for a change in [CO2] (no feedbacks!):
with alpha = 5.35
A doubling of the 1950 [CO2] would compute to dT =
with T = 281.4 K = 8.4°C as the 1950 mean temperature derived from the linear fit and 5.6705E-08 Boltzmann's sigma constant
dT = 0.73 K
which is a much more plausible result based on
Using a water-vapour feedback amplifier of 1.3  this would give a still modest dT = 0.95 K or °C.
We will compare these trends with those of Trier Petrisberg located in Germany about 30km away asap, and also make an analysis on the variability for different periods.
 Nasa GISS data base: http://data.giss.nasa.gov/cgi-bin/csci/csci_station.py?id=629065900002&data_set=1
 Nasa http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=629065900002&data_set=1&num_neighbors=1
 UAH http://climate.uah.edu/
 Michaels, Patrick & Balling, Robert C. Jr.: The Satanic Gases. Cato Institute 2000. ISBN 1-882577-91-4 & ISBN 1-882577-92-2
 Keeling C.D. et al: Mauna Loa CO2 data
 Erren, Hans: How does CO2 respond to temperature ?
 Ramanathan and Vogelmann ,1997 (http://www-ramanathan.ucsd.edu/publications.html)
Calculations and graphs made by Easyplot 4.0 from Spiral Software
06 Nov 2005: changed modified links  and