TY - DATA T1 - Has poor station quality biased U.S. temperature estimates? Supplementary Information PY - 2014/04/19 AU - Ronan Connolly UR - https://figshare.com/articles/dataset/Has_poor_station_quality_biased_U_S_temperature_estimates_Supplementary_Information/1004025 DO - 10.6084/m9.figshare.1004025.v2 L4 - https://ndownloader.figshare.com/files/3176642 L4 - https://ndownloader.figshare.com/files/1470378 L4 - https://ndownloader.figshare.com/files/1470377 L4 - https://ndownloader.figshare.com/files/1470388 L4 - https://ndownloader.figshare.com/files/1470380 L4 - https://ndownloader.figshare.com/files/1470386 L4 - https://ndownloader.figshare.com/files/1470382 L4 - https://ndownloader.figshare.com/files/1470383 L4 - https://ndownloader.figshare.com/files/1470384 L4 - https://ndownloader.figshare.com/files/1470385 L4 - https://ndownloader.figshare.com/files/1470389 L4 - https://ndownloader.figshare.com/files/1470387 L4 - https://ndownloader.figshare.com/files/1470390 L4 - https://ndownloader.figshare.com/files/3176645 KW - temperature records KW - NOAA NCDC KW - Weather station exposure KW - Climate Science N2 - Supplementary information dataset for the following article: R. Connolly and M. Connolly (2014). Has poor station quality biased U.S. temperature trends? Open Peer Rev. J., 11 (Clim. Sci.), ver 0.1 (non peer-reviewed draft)   Abstract of article  Two independent surveys have found that about 70% of the thermometer stations in the U.S. Historical Climatology Network (USHCN) dataset are currently poorly or badly sited. Previous investigations into how this poor siting has affected estimates of U.S. temperature trends have led to apparently contradictory conclusions. However, in this study, these contradictions are resolved, and it is shown that poor station quality has introduced a noticeable warming bias into temperature trend estimates for the U.S. For the unadjusted station records, this poor siting increased the mean temperature trends by about 32%. When time-of-observation adjustments were applied to the records, this increased temperature trends by about 39%, and so the relative fraction of the trends due to the siting bias decreased. However, the siting biases were still substantial, and increased trends by about 18%. The step-change homogenization algorithm which had been developed to remove non-climatic biases such as siting biases was shown to be seriously problematic. Instead of correcting the poorly- and badly-sited station records to match the trends of the well-sited stations, it appears to have blended the temperature records of all stations to match the trends of the poorly-sited stations. It seems likely that similar poor siting biases also exist in global thermometer datasets, and this has probably led to an overestimation of the amount of “global warming” since the 19th century. ER -