Problems converging ndown

[afernandezody]

Hello,
I have spent the last few days trying to run 'ndown' but it always generates this type of error:

----- ERROR: The reference pressure is not monotonically decreasing
This tends to be caused by very high topography
(i,j) = 1 1 , topography = 0.00000000 m
k = 1 , reference pressure = 5000.00000 Pa
k = 2 , reference pressure = 5000.00000 Pa
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 1249
In the dynamics namelist record, reduce etac from 0.200000003

Decreasing etac to 0.1, 0.05 or even lower makes no difference. My initial domains were relatively large so I tried changing the number of eta levels (between 33 and 51), changing the dataset so that the pressure at the top is 1,000 Pa rather than 5,000 Pa, changing the number of days, and the resolution. After so many failed attempts, I have stripped the model and decreased the domain to a minuscule area next to the sea with no mountains. The error message has changed slightly

Input data is acceptable to use: wrfout_d01_2026-07-11_00_00_00
----- ERROR: The reference pressure is not monotonically decreasing
This tends to be caused by very high topography
(i,j) = 1 1 , topography = 5.19905329 m
k = 1 , reference pressure = 4938.74219 Pa
k = 2 , reference pressure = 5000.00000 Pa
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 1249
In the dynamics namelist record, reduce etac from 0.200000003

yet it is still crashing. WRF (and obviously ndwon) is v4.8 and WPS is v4.7 so I was wondering if anyone is running this combination, and has any suggestion. I'm attaching the namelists of the last simulation just in case anyone wants to take a peak.
Thanks.
P.S. After some digging, I think that the issue originates when calling compute_vcoord_ld_coeffs from vert_cor. However, I still don't know how to fix it.