Let us now test my guesses against Sirius:

Ana-mua, entrance pillar	Antares, α Scorpii	16h 26
Ana-roto, middle pillar	Spica, α Virginis	13h 23
Te Pou	Sirius, α Canis Majoris	06h 43
Ana-muri, rear pillar (at the foot of which was the place for tattooing)	Aldebaran, α Tauri	04h 33

Sirius is rising after Alderbaran, just like Spica. 6 * 60 + 43 = 360 + 43 = 403 minutes. Aldebaran has 4 * 60 + 33 = 240 + 33 = 273 minutes. Remarkably we have 360 at Sirius and 240 at Aldebaran, numbers we are familiar with. They are roughly 2 hours apart, and with 24 hours in a day and 12 months in a year it means they are roughly 1 month apart. 360 minutes is a cycle if we regard a quarter of a year as a cycle.

403 - 273 = 123 minutes, which we can transform into glyphs by dividing by 3.94.

123 / 3.94 = 31 and with Aldebaran at 186, Te Pou should be around 186 + 31 = 217.

a1	30	30	b1	26	26
a2	29	59	b2	35	61
a3	24	83	b3	30	91
a4	27	110	b4	33	124
a5	30	140	b5	29	153
a6	29	169	b6	28	181
a7	34	203	b7	31	212
a8	26	229	b8	30	242
sum	229		sum	242

217 is the ordinal number for Ga8-13:

Ga7-32	Ga7-33	Ga7-34	Ga8-1
Ga8-2	Ga8-3	Ga8-4
Ga8-5	Ga8-6	Ga8-7	Ga8-8	Ga8-9
Ga8-10	Ga8-11	Ga8-12
Ga8-13 (217)	Ga8-14	Ga8-15

The kuhane station Te Pou is at 9 * 29.5 = 265.5, which is rather far away.

Ga8-13 does not seem very convincing, although 9 * 24 = 216 might be relevant (266 - 216 = 50). Maybe neither mago at Ga7-16 nor vaha mea at Ga1-4 refer to stars. I am more inclined to rely on 266 as equal to the position of Sirius than to rely on a supposed equality between Ga7-16 and Aldebaran.

But then we can try it the other way around. Suppose Sirius is at 266, what position would then Aldebaran have? Easy piece: 31 glyphs earlier. 266 - 31 = 235, close to Te Pei:

Ga8-16	Ga8-17	Ga8-18	Ga8-19	Ga8-20
Ga8-21	Ga8-22	Ga8-23	Ga8-24	Ga8-25
Ga8-26	Gb1-1	Gb1-2	Gb1-3	Gb1-4
Gb1-5 (235)	Gb1-6	Gb1-7	Gb1-8	Gb1-9

And Antares would be around 235 - 182 = 53, i.e. at Ga1-22:

Ga1-17	Ga1-18	Ga1-19	Ga1-20	Ga1-21	Ga1-22 (53)	Ga1-23	Ga1-24	Ga1-25

If we count from winter solstice instead of from Gb8-30, Antares will be at 64 + 53 = 117 (close to 4 * 29, indeed exactly 4 * 29 if we count Gb8-30 only once). Aldebaran will be at 64 + 235 (where we can read 235 as 53 backwards preceded by 2) = 299, and Sirius at 64 + 266 = 330.

Winter solstice is 6h away from spring equinox. Sirius is not far from 6h, and it will be not far from 12h if we count from winter solstice (north of the equator). Sirius must rise in summer north of the equator. On Easter Island it will rise not far from winter solstice.

Sirius is at 6h 43 = 403 minutes. If we raise with 360 minutes (6 h), it will become 403 + 360 = 763. 12h = 720 and 763 - 720 = 43 minutes, equal to 43 / 3.94 = 11 glyphs. 330 + 11 = 343. But winter solstice should be, we have earlier found, around Gb3-12:

Gb3-12 is rather the 1st day after solstice, though, because we counted Gb8-30 only once when we arrived at 63 + 303 = 366. A more correct glyph for winter solstice is Gb3-11:

Gb2-27	Gb2-28	Gb2-29	Gb2-30	Gb2-31
Gb2-32	Gb2-33	Gb2-34	Gb2-35	Gb3-1
Gb3-2	Gb3-3	Gb3-4 (295)	Gb3-5	Gb3-6
Gb3-7	Gb3-8	Gb3-9 (300)	Gb3-10	Gb3-11 (302)

At winter solstice a generative power is at work. Gb3-10 has 8 + 8 = 16 feathers, and in Gb3-9 we recognize the 'swallower of sun' (in contrast to the sun swallow which announces summer), adequately at position 10 * 30 = 300.

The 10th kuhane station, Hua Reva, is beginning at Gb3-4, sun is swallowed 5 days later, and winter solstice comes yet another 2 days later.

This is fact, not fiction, I think. Sirius ought then to arrive after Gb3-11 and not before. However, according to the text of G it is located at 266, which is 36 (!) days earlier.

Presumably we can define Aldebaran (Ana-muri) as positioned close to where the reader of side a must turn over the tablet in order to continue on side b:

Gb1-5 (235)	Gb1-6	Gb1-7
'October 15'	'October 16'	'October 17'

235 + 64 = 299 indicates zeroes are on their way. The situation resembles what is described 29 glyphs earlier:

Ga7-15	Ga7-16 (186)	Ga7-17
248	249	250
'August 24'	'August 25'	'August 26'

Maybe we now can begin to understand the enigmatic upraised arm in Gb1-5, it resembles the right part of the open jaw of mago. It resembles the shape of waning moon (south of the equator).

By turning the tablet a new season will arrive. Although the text begins at bottom left on the tablet (exactly as on  side a) the reality is that sun now is moving down, not up. The text flow is opposite to the movement of the sun.

With Te Pou located before winter solstice in the text (as read by the kuhane), it must be a mirror image of the day-side 'reality', where Sirius comes after winter solstice. Sirius is at 6h 43 = 403 minutes, 43 minutes after winter solstice, equal to 11 glyphs after solstice.