Investigation of temperature variations in road pavements under tropical climatic conditions
MetadataShow full item record
The purpose of this research study was to record / actual pavement temperatures at different levels within a rigid (concrete) and flexible (bituminous) pavement slabs under tropical climatic conditions. Generally, however, road temperatures are known to be influenced by several factors, such as: i) thermal properties of the pavement materials, notably, thermal conductivity, thermal diffusivity, thermal absorptivity and thermal emissivity; ii) heat flow into the ground; iii) prevailing air and surface temperatures; iv) general meteorological factors such as: a) rainfall, evaporation and humidity, b) cloud cover, c) solar and net radiation, d) wind speed. It was apparent, therefore, that not only did this investigation involve measurements of slab temperatures at different levels, but also s imult an ecus recording of some of the factors that influence road pavement temperatures. Thus air temperature, solar and net radiation, and heat flow were also recorded. The data collected was used to establish some temperature pattern and correlations between different parameters and slab temperatures. The investigation was carried out as follows: i) Site preparation ii) Instrumentation iii) Data collection iv) Analysis and results. 1 2 An experimental site was selected very close tp Wilson Airport, in the outskirts of Nairobi, not only because of its suitable environmental conditions but also because use could be made of some of the meteorological data which has been recorded there. Two 3 x 3 x 0.15m slabs (one concrete and the other bituminous) were cast on a 15 x 15 m prepared murram subgrade embankment. The concrete slab was cast using a mix which was designed according to British Standard Specification Within the slab, a 25 $ x 150 mm cylinder in which the temperature monitoring copper-constantan thermocouples were incoporated at 0, 50, 100 and 150 mm depths was centrally placed. Underneath the slab, that is, in the subgrade,thermocouple probes were at 0.1+5 and 1.05 metre depths. The casting and instrumentation of this slab was completed in November 1970. In casting the bituminous slab, 3/1+ (l9mm)-inch standard road Premix from the City Council of Nairobi was used. A level was used to accurately place (centrally) the temperature measuring thermocouples at 0.00, 0.05, 0.10, 0.15, 0.30 , 0.1*5 and 1.05 metre depths Casting and instrumentation of this slab was completed %H. in March 1971* Including the air temperature measuring thermocouple, which was placed in a standard Stevenson screen at a height of 1.2m above the slab top surfaces, ll+ thermocouple probes were used. These were connected to a l6-channel Honeywell recorder. Other instrumentation included setting up a solar radiometer, a net radiometer for each slab, a 0.20 x 0.15m heat flow plate on each slab top surface and a cup anemometer at 1+.0m above top surface. Connections from these measuring devices were connected to another l6-channel recorder to enable automatic 3 measurements of solar radiation, net radiation, heat flow and wind srtearl yponortivslv. x“ r «7 All intended records were in form of strip charts from the multipoint recorders, printing out at approximately four minute intervals. Of about two years' temperature data recorded, one year data (April 1971 - March 1972) was extracted and converted to digital output on punched cards in order to facilitate most of the intended analysis by the University's computer, ICL 1902A. Owing to some unforceable instrument problems which could not be corrected in time, only about Uo days of solar, net radiation and heat flow data was collected; and no wind speed measurements were recorded. Daily maximum, minimum and mean temperatures were established for each of the ll+ probe levels, and these together with some hourly temperatures were plotted in order to indicate temperature - time patterns. These temperature-time variation patterns were found to be similar from day to day, and from month to month. Air and slab top surface minimum temperatures occurred at about 6 hours, while maximum temperatures for air, concrete slab top and bituminous top surface occurred generally at 15, 1^, and 13 hours after mid-night •m. respectively. Both concrete and bituminous slab bottom surface minimum temperatures occurred at 8 - 9, and maximum at 17 - 18 hours after mid-night. It was also noted that bituminous slab temperatures at every level were higher than those of concrete slab at corresponding levels; and that this difference narrowed as the depth increased. Daily temperature ranges for air and slab top and bottom surfaces as well as daily maximum (daytime) and minimum (night-time) temperature gradients ^between top and bottom surfaces were evaluated. h Some of these data were plotted in or4er to give graphicalpresentat ion of the daily temperature range and gradient patterns. The extracted data were further used to establish frequency distributions for i) Hourly temperatures ii) Daily maximum temperatures iii) Daily mean temperatures iv) Daily temperature ranges v) Daily maximum temperature gradients between top and bottom faces of the slabs. Results of these frequency distribution analyses were used to plot frequency histograms in each case; and to determine the most likely magnitudes for each variable as well as the percentage of the year within the most probable valued- In order to convert the data collected to long term usage and to be able to predict road pavement temperatures from air temperature recordings for which long-term data are generally available, two major relationships were established. Firstly, air-top surface temperature relationship, and secondly; temperature-depth relationship. In the first case, a linear relationship for both hourly and daily mean temperatures was found to exist and the coefficients of correlation (over 90%) were good. In the second case, both maximum and minimum temperature-depth relationships were of hyperbolic form which were transformed to linear relationships by using reciprocal temperature values. The resulting coefficients of linear correlation (over 95%) were also good. 5 Due to the lack of adequate data, the effect of solar radiation, net radiation and heat flow on slab and air temperatures was only very briefly studied. The analysis done showed that a linear relationship (with a correlation of 93-6%) existed between air and solar radiation. On contrary, with the slab temperatures, the coefficients of correlations were generally low (about 50%). Further, in case of the slabs, the effect of these meteorological factors was found to be significant only on slab top surfaces. It is regretted that during this investigation, some of the information that would have led to a more complete analysis of thermal variations and effects in road pavements were not collected. For instance, no thermal strains/stresses, thermal cracks and moisture variations were recorded. Moreover, no laboratory tests were carried out to establish the necessary physical and thermal properties of the slab materials. Owing to these short-comings, it was not possible to evaluate thermal induced stresses or to test the applicability of the existing empirical formulae. Further, only slabs of one constant depth (150 mm) were used in" this study, and therefore, the effect of different slab depths could not be investigated. It is therefore suggested that further investigation should be carried out. This should, preferably, involve slabs of varying dimensions and if possible in different localities within the same climatical region. The general layout should include facilities to record all the necessary factors that influence road pavement temperatures, as well as actual measurements of thermal effects. It is also suggested that all measurements be recorded simultaneously; and that thermal and physical properties of the slab materials be established in the laboratory.
SponsorhipUniversity of Nairobi
Facult of Science, University of Nairobi
Master of Science Thesis