About Nasa DeltaT algorithms recalibration

Why did I recalibrate the algorithms used by NASA computing DeltaT value for years preceeding 1 BCE?
Basically because a number of ancient reports of total or annular solar eclipses exist and some don’t match with the DeltaT value computed here.
The DE431 ephemeris has been made available since April 2013, so it is possible to proceed with an accurate calibration of the DeltaT value based on the written evidence of ancient solar eclipses, of those deemed reliable.
I therefore collected a certain amount of reports of ancient eclipses based on articles that I found on the Internet. I want to mention some:

• G. Henriksson – The acceleration of the Moon and the Universe. The Mass of the Graviton (Advances in Astrophysics Vol. 2 No 3, August 2017)
• G. Henriksson – From the Fifth Oxford International Conference at Santa Fe. (1996)
• Stephenson, Morrison, Hohenkerk – Astronomical dating of seven classical greek eclipses. (Nature vol. 338 – March 1989)
• Boris Banjevic – Ancient eclipses and dating the fall of Babylon – Publ. Astron. Obs Belgrade no. 80 (2006), 251-257

These are what I call the Calibration Points (CP), that I used so far.
As soon as I can read a book that I purchased which lists the ancient recordings of Sumerian solar eclipses I will be able to add other calibration points.

(CP01) – TSE visible from Eshnunna (Iraq) on 3653 October 23 BCE.
(CP02) – TSE visible from Knowth (Ireland) on 3245 October 31 BCE.
(CP03) – TSE visible from Knowth (Ireland) on 3145 January 14 BCE.
(CP04) – TSE visible from Eshnunna (Iraq) on 2929 May 23 BCE.
(CP05) – TSE visible from Khafaje (Iraq) on 2654 August 10 BCE.
(CP06) – TSE visible from Babylon on 2471 April 1 BCE.
(CP07) – TSE visible from Babylon on 1859 May 14 BCE.
(CP7A) – TSE visible from Ekenberg (Sweden) on 1596 Mar 18 BCE.
(CP08) – TSE visible from Ugarit on 1375 May 3 BCE.
(CP09) – TSE visible from Anyang (China) on 1330 June 14 BCE.
(CP9A) – TSE visible in Ekenberg (Sweden) on 1230 Jul 18 BCE.
(CP10) – TSE visible from Ugarit on 1223 March 5 BCE.
(CP11) – TSE visible from Thebes (Greece) on 648 April 6 BCE.
(CP12) – PSE visible from Thasos (Greece) on 588 July 29 BCE.
(CP13) – TSE visible from Thasos (Greece) on 585 May 28 BCE.
(CP14) – PSE visible from Thebes (Greece) on 488 September 1.
(CP15) – ASE visible from Thebes (Greece) on 478 February 17.
(CP16) – TSE visible from Thasos (Greece) on 463 April 30.
(CP17) – ASE visible from Kultepe (Turkey) on 431 August 3.
(CP18) – ASE visible from Thasos (Greece) on 394 August 14.
(CP19) – TSE visible from Chaeronea (Greece) on 310 August 15.

For being total or annular at the selected locations these eclipses have a minimum and a maximum range for DeltaT computed by means of DE431 long term ephemeris.
The Tidal Acceleration used by DE431 is (−25.858±0.003″/century2) and DeltaT is expressed in seconds.
I calculated the following values of DeltaT Range for TSE and ASE only, then the computed NASA and Recalibrated values.

CP       DeltaT Range      NASA      RC
(CP01) - 94400 to 99500    97630     97630
(CP02) - 82600 to 83750    81663     83627
(CP03) - 80900 to 81900    78495 *   80388 *
(CP04) - 72050 to 74100    71800 *   74020
(CP05) - 61700 to 67750    63712     65797
(CP06) - 59400 to 63300    58612 *   60338
(CP07) - 42700 to 45400    43072     43177
(CP7A) - 34450 to 38950    37132     37525
(CP08) - 33500 to 34600    32474 *   32862 *   was it total?
(CP09) - 31800 to 32250    31563 *   32054
(CP9A) - 29550 to 33400.   29587     30848
(CP10) - 30650 to 31470    29457 *   30783
(CP11) - 18430 to 20020    19363     19116
(CP13) - 17650 to 19620    18384     18100
(CP15) - 15500 to 17200    16717     16393
(CP16) - 13200 to 17000    16457     16177
(CP17) - 15000 to 15800    15923 *   15717
(CP18) - 14000 to 23800    15342     15197
(CP19) - 14000 to 16750    14127     14076

Failure to predict TSE or ASE is marked with an asterisk. Nasa’s algorithms have some problem here.
CP03 needs further investigation about totality. I’ve not found any other info about Knowth’s eclipses in the Internet. Mentioned by G. Henriksson only.

I wonder if CP08 was really a total eclipse in Ugarit.
In 1948 a clay tablet (KTU 1.78) was discovered among the ruins of the Royal Palace in Ugarit.
The tablet was translated and an evidence of a solar eclipse reference was found.
The translated text goes: “On the day of the new moon in Hiyaru (month’s name) the Sun went down, its gate keeper was Rsp (Mars). Two livers examined : danger.“
In the 70s this report was attributed to the solar eclipse that took place on 1375 BCE, May 3, likely seen as total in Ugarit.
But there was something wrong with Mars, it was not visible at all during this eclipse, the planet being 6° below the horizon.
The attribution of the eclipse to 1375 BCE was probably made only because of a misinterpretation of the month Hiyaru, and the lack of presence of Mars ignored.
However in 1989 an article published in the journal Nature by T. de Jong and W.H. Van Soldt proposed as the subject of the report the total solar eclipse that took place in 1223 BCE, March 5, bringing justice to Mars that was actually 3.5° distant from the event.
That said it’s clear that CP10 was total in Ugarit because Mars could be visible only during totality and CP08 most likely was not total in Ugarit due to the DeltaT range computed by means of DE431.

Anyway I still have to add more calibration points to get better results.
A final consideration: the livers were right. After no more than 30 years this eclipse took place, Ugarit was sacked by the Sea Peoples and disappeared as a settlement.