Fig. 118. Ten-year-long (1994-2003) temperature monitoring series in Sporilov hole. Plot shows temperatures that were measured at the surface and at 38.3 m depth. Line superimposed on the surface temperature series represents an estimate of the linear trend.
0.3-4K, strongly depending on locality. The fact is that this warming was not derived from the SAT records. For example, Karl et al. (1991), after analyzing the meteorological station records for the mid-continent, concluded an absence of statistically significant climatic trends. As shown in Figure 118, the amplitude of the surface temperature variations does not increase with the overall trend. This means that the variance is not correlated with the mean over the segments of the series (see Section 4.3.1).
Detected warming trend is illustrated in more detail in Figure 119 that compares temperatures monitored during 1994-2005 in Sporilov hole at 38.3 m depth with annual average warming rates. For the decade, temperature has warmed from 10.63°C in 1994 to 10.89°C in 2003. The monitoring results exhibit closely parallel linear trends for the individual years for the period from 2000 to 2005 and a progressive rise of the warming rate from 0.0296K/year in 1994 to 0.0402K/year in 2003. This warming was not one-way story. Warming was stronger in the year 1996 than in the years 1997-1999. The greatest warming rate of the whole 1994-2005 observational period has occurred in the year 2002. Smaller and less significant mean warming rate of only 0.026 K/year reflects more complex course of the temperature increase on decadal scale. Because of attenuation of high frequencies, trends at all depths in the underground have the same or even 2-3% lower relative error than those calculated from the data monitored in the air. In other words, subsurface trends are determined with the same or little bit higher accuracy as the SAT trends.
As the subsurface is seeing more remote events, the amounts of the surface and deeper-measured trends and their timing cannot be compared directly. Figure 120 illustrates the penetration of the linear warming trend occurred at the surface to the depth. This process can be described by Eqs. (2.14) and (2.15) (Section 2.3.3, Chapter 2)
with n = 2. Velocity of penetration depends on the thermophysical properties of the medium and not on the rate of the surface warming. The fingerprint of the surface warming is already measurable at 1m depth after ~20 days from the beginning of the surface warming, after ~220 days at 10m depth, and after 3 and ~7 years at 30 and 50m depth, respectively. In the case of sustained warming, an amount of the linear trend observable, e.g., at 10m depth achieves 70% of the surface value after 10-14 years from the beginning of warming. At 30m depth, warming rate will achieve 50% of the surface value after 25-40 years from the beginning of warming. Comparing the trends detected by Sporilov monitoring experiment with the long-term SAT record at meteorological station Prague-Klementinum (Figure 64, Chapter 2), it can be concluded that today's subsurface likely reflects strong warming trend that began in the area after the relatively cold 1940s (see also Cermak et al., 2000). As was shown by both the GST reconstructions and the analysis of the SAT data in the territory of the Czech Republic, this warming trend is characteristic for the wide territory surrounding Prague (Section 3.1, Figure 82). An independent analysis of the SAT records from 30 Czech meteorological stations (period 1961-1996) has revealed warming trends that fall in the interval from 0 to 0.04K/year with characteristic regional warming rate of 0.0283 K/year (Cermak et al., 2000). Approximately 60% of the results fall within 0.02-0.03 K/year interval. An analysis of the spatial pattern of this trend has confirmed the conclusion by Bodri and Cermak (1999) that more pronounced recent warming is observed in more populated and generally industrialized areas, while lower values occur in either agricultural or forested areas.
Fig. 120. Penetration of the linear warming trend into the subsurface. Curves are labeled by the depth level. After 10-14 years the warming at 10 m depth will achieve about 70% of the value of the surface warming (k = 10"6m2/s).
Because at least a part of the warming observed at Sporilov can be attributed to an anthropogenic contribution to the local climate in a large urban agglomeration (so-called "urban heat island" effect3), similar monitoring experiment was performed in the Kocelovice site, the Czech Republic (49.47°N, 13.84°E, 518m a.s.l.; Cermak et al., 2000). The locality represents a rural zone. The 40 m deep borehole is situated at the territory of the meteorological station, and monitoring experiment was put in operation in 1998. Thermistor sensors were fixed in depths of 0.02, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 4, 5, 7.5, 10, 15, 20, 25, 30, 35, and 40m. Air temperature was measured at 0.2, 1, and 2 m. In addition to the temperature measurements, the level of underground water, precipitation, snow cover thickness, wind speed and direction, solar radiation, and air moisture were also registered. The rate of registration was once in an hour. Figure 121 shows the year 2003 temperature increase recorded in the Kocelovice borehole. Detected warming rate was 0.0176 K/year in 1999 and thus was only near 70% lower than that
3An urban heat island effect (UHI) corresponds to significantly warmer urban agglomeration area than its surrounding countryside. The principal reasons for the UHI are the comparatively warm buildings, significantly differing thermophysical properties of the surface materials used in urban areas (like as asphalt; see results of the monitoring experiments described in Section 2.6.2), and the lack of evapotranspiration (Section 2.6.3). The process of the population agglomeration growth is generally accompanied by a corresponding increase in average temperature that in principle can be confused with the warming trend occurring due to global warming phenomenon.
Was this article helpful?