Reassessing changes in diurnal temperature range: A new data set and characterization of data biases

Abstract

It has been a decade since changes in diurnal temperature range (DTR) globally have been assessed in a stand-alone data analysis. The present study takes advantage of substantively improved basic data holdings arising from the International Surface Temperature Initiative's databank effort and applies the National Centers for Environmental Information's automated pairwise homogeneity assessment algorithm to reassess DTR records. It is found that breakpoints are more prevalent in DTR than other temperature elements and that the resulting adjustments have a broader distribution. This strongly implies that there is an overarching tendency, across the global meteorological networks, for nonclimatic artifacts to impart either random or anticorrelated rather than correlated biases in maximum and minimum temperature series. Future homogenization efforts would likely benefit from simultaneous consideration of DTR and maximum and minimum temperatures, in addition to average temperatures. Estimates of change in DTR are relatively insensitive to whether adjustments are calculated directly or inferred from adjustments returned for the maximum and minimum temperature series. The homogenized series exhibit a reduction in DTR since the midtwentieth century globally (À0.044 K/decade). Adjustments serve to approximately halve the long-term global reduction in DTR in the basic "raw" data. Most of the estimated DTR reduction occurred over 1960-1980. In several regions DTR has apparently increased over 1979-2012, while globally it has exhibited very little change (À0.016 K/decade). Estimated changes in DTR are an order of magnitude smaller than in maximum and minimum temperatures, which have both been increasing rapidly on multidecadal timescales (0.186 K/decade and 0.236 K/decade, respectively, since the midtwentieth century).

Shifting Hardiness Zones: Trends in Annual Minimum Temperature

“…In general, global studies of surface temperatures have found daily minimum temperatures to be increasing faster than daily maximums and winter temperatures to be increasing faster than summer temperatures, resulting in reduced diurnal and seasonal temperature ranges [24][25][26][27]. The most recent Intergovernmental Panel on Climate Change report concluded with medium confidence that globally, daily minimum temperatures have increased more than daily maximum temperatures since 1950 [28].…”

Spatial and temporal variations in extreme temperature in Central Asia

“…The same trends were also observed in Saudi Arabia from 1981 to 2010 (Nazrul Islam et al, ), the mountain stations in the Koshi River Basin (Shrestha et al, ) from 1901 to 2003 (Rai et al, ) and Serbia from 1981 to 2010 (Ruml et al, ). These results were seemingly consistent with the trends in DTR that have been observed since 1979, when global changes in DTR became insignificant (Thorne et al, , 2016b). However, in Xinjiang (China), due to the significant increase in TMIN, DTR decreased significantly on an annual scale, particularly during summer and winter.…”

The EUSTACE global land station daily air temperature dataset

“…Our results are qualitatively in agreement with those obtained with the Pairwise Homogeneity Assessment algorithm (Thorne et al ., ); in particular, we find more frequent breakpoints (i.e. shorter homogeneous periods) for DTR than for the other variables.…”

“…In global datasets, the large amount of data limits the usability of most of these methods. The Pairwise Homogeneity Assessment (Menne and Williams Jr, ) is one of the few examples of breakpoint detection methods that have been applied to global datasets (e.g., Lawrimore et al ., ; Dunn et al ., ; Thorne et al ., ).…”