A paper that appeared in EOS in October 2005 (subscription required) entitled “Urban Heat Islands Developing in Coastal Tropical Cities” by González et al is very relevant to the subject of long term surface temperature trend analyses such as reported by GISS and other groups. This paper indicates that the trends in surface temperatures in growing tropical urban locations are quite large. The paper states with respect to San Juan, Puerto Rico that,
“A recent climatological analysis of the surface temperature of the city has revealed that the local temperature has been increasing over the neighboring vegetated areas at a rate of 0.06 C per year for the past 30 years. This is a trend that may be comparable to climate changes induced by global warming”.
This study contradicts that of the Parker 2004 Nature paper entitled “Large-scale warming is not urban” and the Journal of Climate regional study of Peterson 2003 entitled “Assessment of urban versus rural in situ surface temperatures in the contiguous US: No difference found”. These later two studies were applied in the 2006 CCSP Report “Temperature Trends in the Lower Atmosphere:
Steps for Understanding and Reconciling Differences” to demonstrate the robustness of the global surface temperature trend analyses. The neglect of the conclusions of the González et al EOS paper (which was published prior to the completion of the CCSP Report), represents yet another example of its completion as an advocacy document in support of a particular limited perspective on surface and tropospheric temperature trends (see also my Public Comment on this CCSP Report).
Moreover, since such a large urban trend, if representative of other coastal tropical cities, instead of meaning that “This is a trend that may be comparable to climate changes induced by global warming”, indicates that the “global warming” signal itself, as diagnosed by surface temperatures, has a warm bias, when interpolated to a larger grid analysis area as a result of these large urban temperature trends.