## Diurnal variation of solar irradiance

For a given set of atmospheric conditions, the irradiance at any point on the Earth's surface is determined by the solar elevation, b. This rises during the day from zero (or its minimum value in the Arctic or Antarctic during the summer) at dawn to its maximum value at noon, and then diminishes in a precisely symmetrical manner to zero (or the minimum value) at dusk. The exact manner of the variation of b with time of day depends on the latitude, and on the solar declination, d, at the time. The declination is the angle through which a given hemisphere (north or south) is tilted towards the Sun (Fig. 2.5). In summer it has a positive value, in winter a negative one; at the spring and autumn equinoxes its value is zero. Its maximum value, positive or negative, is 23 °27'. Published tables exist in which the solar declination for the northern

21 March

21 March 22 December

Northern winter

Fig. 2.5 Variation of the solar declination throughout the year.

22 December

Northern winter

Fig. 2.5 Variation of the solar declination throughout the year.

hemisphere may be found for any day of the year. Alternatively, a relation derived by Spencer (1971) may be used d = 0.39637 - 22.9133 cos C + 4.02543 sin C - 0.3872cos 2C + 0.052 sin 2C

where C is the date expressed as an angle (C = 360° x d/365; d = day number, ranging from 0 on 1 January to 364 on 31 December): both d and C are in degrees. The declination for the southern hemisphere on a given date has the same numerical value as that for the northern hemisphere, but the opposite sign.

The solar elevation, b, at any given latitude, g, varies with the time of day, t, (expressed as an angle) in accordance with the relation sin b = sin g sin d — cos g cos d cos t

where t is 360° x t /24 (t being the hours elapsed since 00.00 h). If we write this in the simpler form sin b = c — c2 cos t

where c1 and c2 are constants for a particular latitude and date (c1 = sin g sin d, c2 = cos g cos d), then it becomes clear that the variation of sin b with the time of day is sinusoidal. Figure 2.6 shows the variations in both b and sin b throughout a 24-hour period corresponding to the longest summer day (21 December) and shortest winter day (21 June) at the latitude of Canberra, Australia (35 ° S). For completeness, the values of b and sin b during the hours of darkness are shown: these are negative and correspond to the angle of the Sun below the horizontal plane. Fig. 2.6 Variation in solar elevation (b) and sin b during a 24-h period on the longest summer day and the shortest winter day at 35° latitude.

The variation of sin b is sinusoidal with respect to time measured within a 24-hour cycle, but not with respect to hours of daylight.

The sine of the solar elevation b is equal to the cosine of the solar zenith angle 0. Irradiance due to the direct solar beam on a horizontal surface is, in accordance with the Cosine Law, proportional to cos 0. We might therefore expect that in the absence of cloud, solar irradiance at the Earth's surface will vary over a 24-hour period in much the same way as sin b (Fig. 2.6), except that its value will be zero during the hours of darkness. Exact conformity between the behaviour of irradiance and sin b is not to be expected since, for example, the effect of the atmosphere varies with solar elevation. The greater attenuation of the direct beam at low solar elevation due to increased atmospheric pathlength, although in part balanced by a (proportionately) greater contribution from skylight, does reduce irradi-ance early in the morning and towards sunset below the values that might be anticipated on the basis of eqn 2.8. This has the useful effect of making the curve of daily variation of irradiance approximately sinusoidal with respect to daylight hours, even though, as we have seen, the variation of sin b is strictly speaking only sinusoidal with respect to hours since 00:00 h. Figure 2.7 gives some examples of the diurnal variation of total irradiance at different times of year and under different atmospheric conditions.

A smooth curve is of course only obtained when cloud cover is absent, or is constant throughout the day. Broken cloud imposes short-term irregularities on the underlying sinusoidal variation in irradiance (Fig. 2.7). If E(t) is the total irradiance at time t h after sunrise, then by integrating E(t) with respect to time we obtain the total solar radiant energy received per unit horizontal area during the day. This is referred to as the daily insolation and we shall give it the symbol, Qs

Qs where N is the daylength.

If E(t) has the units W m~2 and N is in s, then Qs is in J m~2. On a day of no, or constant, cloud cover the diurnal variation in E(t) is expressed approximately by

where Em is the irradiance at solar noon.928 By integrating this expression in accordance with eqn 2.9, it can be shown that the daily insolation on such a day is related to the maximum (noon) irradiance by

Thus if, as for example in Fig. 2.7c (16 March), the maximum irradiance is 940 W m~2 in a l2-hour day, then the total energy received is about 26 MJ.

On days with broken cloud, the degree of cloud cover is varying continuously throughout any given day. Nevertheless, in many parts of the world, the cloud cover averaged over a period as long as a month is approximately constant throughout the day.1399 Therefore, the diurnal variation in total irradiance averaged over a month will be approximately sinusoidal and will conform with eqn 2.10. From the average maximum irradiance, the average daily insolation can be calculated using eqn 2.11.

Solar

Solar Fig. 2.7 Diurnal variation of total solar irradiance at different times of year and under different atmospheric conditions. The measurements were made at Krawaree, NSW, Australia (149°27' E, 35°49' S; 770m above sea level) on dates close to (a) the shortest winter day, (b) the longest summer day, (c) the autumn equinox. (-clear sky; intermittent cloud; generally overcast). Data provided by Mr F.X. Dunin, CSIRO, Canberra.

Time (Australian Eastern Standard Time)

Fig. 2.7 Diurnal variation of total solar irradiance at different times of year and under different atmospheric conditions. The measurements were made at Krawaree, NSW, Australia (149°27' E, 35°49' S; 770m above sea level) on dates close to (a) the shortest winter day, (b) the longest summer day, (c) the autumn equinox. (-clear sky; intermittent cloud; generally overcast). Data provided by Mr F.X. Dunin, CSIRO, Canberra.