Lo que más mola y más científico queda es que ni ponen barras de error y que comparan los datos de 1880 con los de ahora.Como había tantas estaciones,tan fiables y que medían bien...Es una batalla perdida,que sigan con sus gráficas y mapas de colorines,si son felices...
GISS , publica el margen de error por falta de cobertura espacial , que se calcula al principio de la serie en aprox. +/-0,1C se ve en la barra de color verde de la sig, imagen
Pero no agrega la incertidumbre estructural asociada a su método para calcular la temperatura global por ejemplo la ( extrapolacíon) y a la medición en si de las estaciones met .. solo asume que sumadas todas , seria el doble que la primera
The GISS results are really estimates based on the available data. Accurate error estimates are hard to obtain. However, it is likely that the largest contribution to the margin of error is given by the temporal and spatial data gaps. That particular margin was estimated as follows: All computations were first made replacing the observed data by complete model data. Then the calculations were repeated after discarding model data where the corresponding observations were missing. Comparisons of the two results were used to obtain an estimate for that margin of error. Assuming that the other inaccuracies might about double that estimate yielded the error bars for global annual means drawn in this graph, i.e., for recent years the error bar for global annual means is about ±0.05°C, for years around 1900 it is about ±0.1°C. The error bars are about twice as big for seasonal means and three times as big for monthly means. Error bars for regional means vary wildly depending on the station density in that region. Error estimates related to homogenization or other factors have been assessed by CRU and the Hadley Centre (among others).Pero como dicen los de GISS ,mas arriba ,HADCRUT4 si que contempla todos las posibles fuentes de incertidumbre , y las calcula .. por lo que se puede hacer un simple gráfico de HADCRUT4 con su incertidumbre y compararla con GISS haber que tan lejos o cerca están uno de otro , y si GISS se sale del margen de error de HADCRUT.
Como se ve , GISS no solo no se escapa del margen de error sino que van bastante cerca uno de otro en su estimación central ..
DATOS:
GISS
http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A2.txtHADCRUT
http://www.metoffice.gov.uk/hadobs/hadcrut4/data/current/time_series/HadCRUT.4.4.0.0.annual_ns_avg.txt