This grapher states that the Sun energy output in watts/m2 has not changed in 2200 years (0.0022 Ma ago until 0 Ma ago) .

Why is there an invariance as time approaches glacial minimum?

Years Ago Insolation (W/m2)
2200 342.052 754
2.1 342.052 556
2.0 342.052 362
1.9 342.052 175
1.8 342.051 989
1.7 342.051 802
1.6 342.051 613
1.5 342.051 421
1.4 342.051 225
1.3 342.051 023
1.2 342.050 814
1.1 342.050 598
1.0 342.050 373
0.9 342.050 128
0.8 342.049 875
0.7 342.049 615
0.6 342.049 349
0.5 342.049 079
0.4 342.048 806
0.3 342.048 532
0.2 342.048 257
0.1 342.047 984
0.0 342.047 714
  • 1
    $\begingroup$ This question could be made clearer and more concise. But in essence mean insolation, the average input of solar energy, is not what changes in Milankovitch cycles. These cycles change the eccentricity of our orbit and tilt of the earth's axis. These changes affect the balance of energy falling on the north and south hemispheres, and this in turn is thought to affect climate. $\endgroup$
    – Andy M
    Nov 28, 2022 at 15:03
  • $\begingroup$ @AndyM Surely the orbits eccentricity changes the mean (mean over a day or over an orbit??) insolation of solar energy.. If I take the limit e->0.9999... then I can reach arbitrarily low insolation at the apoapsis.. furthermore most of the orbital time is spent in the far-away region. So while increasing e doesn't increase the orbital period, the average insolation over an orbit certainly decreases. $\endgroup$ Nov 29, 2022 at 0:42
  • 3
    $\begingroup$ @LifeInTheTrees: There's no invariance in the data you show, but a clear trend. Look at the second and third digit after the comma, and compare that to the magnitude of your expectations from the Milankovich cycles. $\endgroup$ Nov 29, 2022 at 0:44

1 Answer 1


Look at the article they cite as the source of their computations. It's a model that's meant to be valid across hundreds of millions of years, only based on orbital parameters (changes in Earth's rotation, axial tilt, distance from the sun, dates of aphelion/perihelion, etc). These are all parameters that only change significantly over the course of tens of thousands of years. The idea is to understand that, due to those orbital changes, some places on Earth may have gotten, on average, more or less sun in the past than they do now (e.g. the current near-alignment of perihelion with the northern winter solstice was opposite at some time in the past, resulting in a different balance of insolation between the northern and southern hemispheres).

What it doesn't model is the actual radiation output of the sun, known as solar constant. Solar output varies over shorter timescales (with significant oscillations having periods between 10 and 1000 years, and some unpredictable dynamics over a scale of months to decades), but this variation isn't included in the model on that page, it's simply averaged out.

  • $\begingroup$ It's an algorythm for milankovitch cycles, i thought there would be 21k year glacial insolation variable, I think that ice age variation of 3-4 Celsius is only axial tilt related. $\endgroup$ Dec 5, 2022 at 11:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.