From CO2Science: Significant urban heat islands have been observed in many cities. To date, much research has revealed urban heat islands are capable of introducing large and spurious warming trends in near-surface air temperature records, which makes it a challenge in determining the true effect of rising atmospheric CO2 on global temperature change. Less studied, however, is the extent to which urbanization impacts land-surface temperature. Yet, ascertaining the influence of urbanization on land-surface temperature is also necessary in order to develop accurate projections of future climate change. This is because land-surface processes impact weather and climate via surface reflectance, soil moisture, soil heat storage and other processes that act to determine the extent of frozen soil and the near-surface heat balance of the atmosphere. Within that context, Bian et al. (2017) studied the effect of urbanization on land-surface temperature in Shijiazhuang, north China. More specifically, they compared and analyzed trends in land-surface temperature (LST) from urbanized Shijiazhuang against two less-urbanized nearby locations (Gaocheng and Yuanshi meteorological stations) over the period 1965-2012.
Paper Studied: Bian, T., Ren, G., Yue, Y. 2017. Effect of urbanization on land-surface temperature at an urban climate station in north China. Boundary-Layer Meteorology 165: 553-567.
Results of the study revealed linear trends in mean LSTs at the urban and rural sites were, respectively, +0.13 °C and -0.14 °C per decade (see Figure 1), resulting in an urban-rural linear trend difference in LST of 0.27 °C per decade. Additional analyses of maximum, minimum and mean LST data by month, season and year indicated the urbanization effect “mainly affects the maximum [land-surface] temperature, in contrast to the larger effect of urbanization on time series of the minimum surface-air temperature at the same urban station” (emphasis added). Further, Bian et al. also report the urbanization contribution to the annual mean urban-rural LST difference (i.e., the 0.27 °C per decade trend) is near 100%, indicating “that the increase in annual mean LST at the Shijiazhuang station is probably due entirely to urbanization.”
The authors conclude their work by stating the significant urbanization effects in the LST time series “indicate a large relative surface warming in urban areas compared with rural areas,” presenting a “problem of data representativeness” in urban stations for “monitoring and studying large-scale climate change.” And until such problems are effectively addressed and accounted for, the true impact of rising atmospheric CO2 on global climate may well be vastly overstated given that up to 100% of mean annual LST in this study was determined in this study to be the result of urbanization, leaving no room for rising CO2.
Figure 1. Anomaly series and linear trends of annual mean land-surface temperature (LST) at the Shijiazhuang urban (solid black lines) and rural (dashed black lines) stations for the period 1965-2012. Source: Bian et al. (2017).
This dovetails nicely with the fact that if you look at temperature trends amongst rural stations in North America you find virtually no temperature gradient over the last 120 years. This is especially true if you stick to the interior of the continent, away from the cyclical effects of the oceans and the ameliorating effects of the Great Lakes. The places I’ve examine in particle are rural sites near McPherson Kansas and near Centre North Dakota. The former is pretty much the geographic center of the US; the latter is the geographic center of North America. Both have virtually flat temperature trends.
Oddly, NOAA goes to great lengths to adjust raw temperature records in order to account for minor differences in time of day measurement and other anomalies. But it doesn’t do any adjustments to account for rural versus urban gradients, the cumulative effect of which — across thousands of stations and across 120 years of data — is likely quite significant.
Totally agree with Mr. Puckering above. Many of the NOAA temperature gauges are located in urban areas surrounded by a sea of asphalt and concrete where daytime and nighttime temperatures will be several degrees higher than surrounding rural areas. The heat island effect is well known and the larger the urban area expands (resulting in more pavement) the greater the effect. My home is 10 miles outside of a large urban area and both daytime and nighttime temperatures (during each season) are consistently 5 degrees cooler at my home than the downtown temperatures.
Roger that, Gus. A couple of years ago I had occasion to drive from my rural home into the centre of a large urban area late at night and then immediately back again on a regular basis, picking up someone working downtown who didn't have a car at the time. It was quite noticeable, particularly in the summer months, that the urban temperature as indicated by my car was invariably several degrees higher than the rural temperature.
I have looked at temperature trends where I grew up in Geauga County Ohio, and where I presently live Fairfax County , Virginia. Neither place shows any significant change in temperature over the last 100 years in the rural areas, but about 1 degree Fahrenheit of warming in the urban areas over the same time span.