The present study was conducted for three successive years in a
sandy soil of Ismailia Agricultural Research Station (IARS). The
abscissas of IARS are Latitude 30˚ 3 30 a d o gitude 3 ˚ 0 E
and its elevation 10.2 meter over the sea level. Three thermal sensors
were placed in the soil at depths of 0.05m, 0.10m, and 0.30m from the
soil surface, to study the reflection effects of variations in the temperature
of air on the variation of temperatures of the soil in the studied depths.
The readings of air temperature (Air-T) were recorded with rate of
one reading/hour, While soil temperature (Soil-T), at any soil depth was
recorded with average of one reading/hour (i.e.24 readings/day). This
means that, at any year, the recorded readings of (Air-T) or (soil-T) were
24 x365=8760 readings.
The monthly averages of soil temperatures of the three depths
throughout the three years were 13.83, 14.41, 16.67, 20.45, 25.10, 29.37,
30.29, 30.30, 28.95, 24.40, 19.86 and 15.41°C for month of Jan., Feb.,
March, April, May, Jun., July, Aug., Sept., Oct., Nov., and Dec.,
respectively. The corresponding air temperatures were 13.69, 13.61,
16.07, 19.41, 23.22, 25.88, 27.78, 27.87, 26.01, 22.85, 18.31, and 14.02°C
for the same previous months. The minimum air and soil temperatures
were recorded at Dec., Jan. and Feb. months, while the maximum were
noticed at July and Aug. months.
An obvious increase in T-soil of different depths was obtained in
the second and third year. Temperatures of 0.05m, 0.10m and 0.30m,
recorded increases of 3.07°C, 2.69°C and 2.83°C at the third year for the
mentioned depths, respectively in respect to the first year. Whereas, the
increments in T-air in the second and third year compare with those of the
first year were 1.17 and 0.37°C, respectively.
The statistical correlation coefficient and regression equations
between air temperature and soil temperature of the three depths were
calculated to study the relationship between air and soil temperatures. The
threshold value (a) of the regression equation was higher in summer and
lower in winter months of the three depths. This was due to the higher
temperature at summer. Moreover, air temperature showed more rapid
change than soil temperature.
The total numbers of temperature readings were 2232 for months,
Jan.,March, May, July, Aug., Oct., and Dec. while the regression rejected
unusual numbers were 94, 123, 98, 56, 64, 199 and 146, respectively. It
was 2016 for Feb. with rejected unusual number of 82. Also, they were 2160 for months April, Jun., Sept. and Nov. with rejected unusual
numbers of 143, 96, 111, and 103, respectively.
Generally, the average fit numbers of temperature which used for
calculating the regression were ranged from 93 to 98% from the total
number of temperatures, which interpreted the highly significant
coefficient of these regression equations and reflect the effective effects
of air temperature on soil temperatures. Thus, soil temperature can be
predicted, to some extent, using the air temperature and these regression
equations. Also, the reported regression equations can be used for
estimating the unrecorded (missing) soil temperatures or another year.