Climate And Climatic Variation

I 41

0.04

50 100 ISO

FIGURE 2.16 Variations of eccentricity, obliquity, precession, and the combination of all three factors (ETP) over the last 800,000 years with their principal periodic characteristics indicated by the power spectrum to the right of each time series (upper diagram). Below is the time series of July solar radiation at 10, 65, and 80°N (expressed as departures from A.D. 1950 values). Note that the radiation signal at high latitudes is dominated by the 41,000 year obliquity cycle whereas at lower latitudes the 23,000 precessional cycle is more significant (after Imbrie et al., 1993b; lower diagram: data from Berger and Loutre, 1991).

50 100 ISO

FIGURE 2.16 Variations of eccentricity, obliquity, precession, and the combination of all three factors (ETP) over the last 800,000 years with their principal periodic characteristics indicated by the power spectrum to the right of each time series (upper diagram). Below is the time series of July solar radiation at 10, 65, and 80°N (expressed as departures from A.D. 1950 values). Note that the radiation signal at high latitudes is dominated by the 41,000 year obliquity cycle whereas at lower latitudes the 23,000 precessional cycle is more significant (after Imbrie et al., 1993b; lower diagram: data from Berger and Loutre, 1991).

Changes in the seasonal timing of perihelion and aphelion result from a slight wobble in the Earth's axis of rotation as it moves around the Sun (Fig. 2.17a). The effect of the wobble (which is independent of variations in axial tilt) is to change systematically the timing of the solstices and equinoxes relative to the extreme positions the Earth occupies on its elliptical path around the Sun (known as precession of the equinoxes) (Fig. 2.17b). Thus, 11,000 yr ago perihelion occurred when the Northern Hemisphere was tilted towards the Sun (mid-June) rather than in the Northern Hemisphere's midwinter, as is the case today. Precessional effects are opposite in the Northern and Southern Hemispheres and the change in precession occurs with a mean period of -21,700 yr (see Fig. 2.16).

Clearly, the effects of precession of the equinoxes on radiation receipts will be modulated by the variations in eccentricity; when the orbit is near circular the seasonal timing of perihelion is inconsequential. However, at maximum eccentricity, when differences in solar radiation may amount to 30%, seasonal timing is crucial. The solar radiation receipts of low latitudes are affected mainly by variations in eccentricity and precession of the equinoxes, whereas higher latitudes are affected mainly by variations in obliquity. As the eccentricity and precessional effects in each hemisphere are opposite, but the obliquity effects are not, there is an asymmetry

FIGURE 2.17 a The Earth wobbles slightly on its axis (due to the gravitational pull of the Sun and Moon on the equatorial bulge of the Earth). In effect, the axis moves slowly around a circular path and completes one revolution every 23,000 yr. This results in precession of the equinoxes (Fig. 2.17b).This effect is independent of changes in the angle of tilt (obliquity) of the Earth, which changes with a period of-41,000 years (from Imbrie and Imbrie,

DEC. 21 TODAY

SEPT. 22

DEC. 21

5500

yr ago

SEPT. 22

11,500

yr ago

MAR. 20

FIGURE 2.17b As a result of a wobble in the Earth's axis (Fig. 2.17a) the position of the equinox (March 20 and September 22) and solstice (June 21 and December 21 ) change slowly around the Earth's elliptical orbit, with a period of ~23,000 yr.Thus 11,000 yr ago the Earth was at perihelion at the time of the summer solstice whereas today the summer solstice coincides with aphelion (from Imbrle and Imbrie, 1979).

between the two hemispheres, in terms of the combined orbital effects, which becomes minimal poleward of -70°. It is also worth emphasizing that the orbital variations do not cause any significant overall (annual) change in solar radiation receipts; they simply result in a seasonal redistribution, such that a low summer radiation total is compensated for by a high winter total, and vice versa (A. Berger, 1980).

MAR. 20

MAR. 20

DEC. 21 TODAY

SEPT. 22

DEC. 21

JUNE 21

JUNE 21

DEC. 21

SEPT. 22

DEC. 21

JUNE 21

It is important to note that the periods mentioned for each orbital parameter (41,000, 95,800, and 21,700 yr for obliquity, eccentricity, and precession, respectively) are averages of the principal periodic terms in the equations used to calculate the long-term changes in orbital parameters. For the precessional parameter, for example, the most important terms in the series expansion of the equation correspond to periods of -23,700 and -22,400 yr; the next three terms are close to -19,000 yr (A. Berger, 1977b). When the most important terms are averaged, the mean period is 21,700 yr, but some paleoclimatic records may be capable of resolving the principal -19,000- and ~23,000-yr periods separately (Hays et al., 1976). Similarly, the mean period of changes in eccentricity is 95,800 yr but it may be possible to detect separate periods of -95,000 and -123,000 yr in long high-resolution ocean core records corresponding to important terms (or "beats" produced by interactions of important terms) in the equation (Wigley, 1976). Eccentricity also has a longer-term periodicity of 412 ka, which has been identified in some marine sedimentary records (Imbrie et al., 1993b). Furthermore, the relative importance of all these periods may have changed over time. For example, the 19 ka precessional and 100 ka eccentricity cycles were more significant prior to -600 ka B.P. (Imbrie et al., 1993b). This is one of the enigmas of the paleoclimatic spectrum; during the last one million years the 100 ka period in geological records increased in amplitude yet over the same interval of time the main period associated with eccentricity shifted to lower frequencies (-412 ka).

Orbital variations may also have significance for climatic variations on much shorter timescales. Loutre et al. (1992) calculated insolation changes over the last few thousand years, resulting from changes in precession, obliquity, and eccentricity. They found statistically significant periodicities in insolation (at 65° N in July) of 2.67, 3.98, 8.1, 18.6, 29.5, and 40.2 yr (Borisenkov et al., 1983, 1985). At other seasons and locations, periodicities of 61, 245, and 830-900 yr are significant. These higher frequency variations are very small in amplitude compared to the orbital changes discussed earlier, but they may nevertheless be important for climatic variability on the decadal to millennial timescale. Interestingly, some of the periodicities in incoming insolation due to orbital effects are similar to those identified in sunspot data (which may relate to solar irradiance changes) so the cumulative effects may be significant for short-term climate variability. This matter has received relatively little attention so far.

Considered together, the superimposition of variations in eccentricity, obliquity, and precession produces a complex, ever-varying pattern of solar radiation receipts at the outer edge of the Earth's atmosphere3. To appreciate the magnitude of these variations and their spatial and temporal patterns, it is common to express the radiation receipts for a particular place and moment in time as a departure (or anomaly) from corresponding seasonal or monthly values in 1950. An example is shown in Fig. 2.18 for the month of July at all latitudes (90° N-90° S) from 0 to 200 ka B.P. (A. Berger, 1979). Of particular interest are the radiation anomalies at high northern latitudes (60-70° N), considered by Milankovitch (1941) to be critical for

3 Values of midmonth insolation receipts for December/January and June/July, at 1000-yr intervals for the last 5 Ma, are given on a diskette accompanying the work of Berger and Loutre (1991).

0 0

Post a comment