Rosema et al: Global cooling of Earth’s surface since 1982

A very interesting paper published in Energy and Environment has found that the Earth’s actual surface (as opposed to the air temperature 1.5-2m above it) has been cooling since at least 1982, according to data from the European METEOSAT satellite observation platform, which is in a geostationary orbit above the equator on the Greenwich Meridian. The authors think this is due to increased cloudiness. They point out that the UAH satellite data processed by John Christy also showed a cooling of the troposphere in the 1980’s until ‘adjustments’ were applied to the original computations. It’s a very readable paper, reprinted from the E&E Journal, available here.

INTRODUCTION
EARS Earth Environment Monitoring is providing Meteosat based water and climate
services. To this end, Meteosat visual and thermal infrared images are received hourly
and processed to hemispheric data fields of surface and boundary layer temperature,
global and net radiation, actual and potential evapotranspiration, as well as
precipitation. These data are then applied in drought monitoring, crop yield forecasting
[1,2] and river flow forecasting systems. Since 2009, EARS is also developing drought
and excessive precipitation insurance [3]. For this purpose a 30 year data base of
hourly Meteosat visual (VIS) and thermal infrared (TIR) images has been composed.
In this study we have used these data to study recent change in planetary temperature.
The subject is topical as during the past decades the fear has developed that by the
burning of fossil fuels and the resulting increase in atmospheric CO2 content, mankind
is seriously affecting climate beyond its natural variability and to an extent that will
appear to be detrimental to plant, animal and human live. This fear has been driven by
time series of temperature measurements at meteorological stations, which show an
overall temperature increase of 0.074 K/decade during the last 100 year, increasing to
0.177 K/decade when considering the last 30 year [4]. These air temperature
measurements are taken at 1.5-2 meter height, in an environment that during the past
century has progressively undergone changes in surface conditions due to
urbanization, road building and drainage. In the Netherlands, for example 10 percent
of the national area currently consists of hardened surface. Such changes may have a
significant effect on local air temperatures. Therefore there is a need to supplement
these near surface local air temperature observations with large scale temperature
observations from satellites.

The calibration of satellite data has long been a thorny issue, and the authors utilise an interesting and to me reasonably plausible method to verify the satellite observations. Two small areas show a warming trend. One is located in S.E. Iraq, where Saddam Hussain took his revenge on the Shiite marsh arabs by draining their homeland between the Tigris and euphrates in the following the first gulf war. The consequent drying of the area has led to a ~10C increase in surface temperature, according to their satellite data. They show that from theory, a very similar magnitude is calculated.

Clearly this is a startling result, and some thought needs to be put into the opposite sign of trend found by this study when compared with other datasets and time series. One thought might be that increased cloudiness would perhaps lead to higher noon and night humidity near the surface which would lift the near surface air temperature measured by the surface stations used by organisations such as the MET Office and NOAA. Another possibility is that the mostly positive adjustments which have applied to near surface air temperature data worldwide might be wrong. .