[Frank The Plumber]

If we look at the variables within this process, we can simplify the possible causes.
The net effect seems to be a vaporization or evaporation of the fuel while in the fuel line conduit.
As root causes here are a few to consider;
Fuel formulation
Size and type of fuel filter.
Type of metal used to construct fuel line.
Diameter of fuel line.
Condition of fuel line.
I would focus my aim towards the condition of the fuel line and the condition of the surface areas of the vessel holding the fuel. Most research is done in a style known as bench, under ideal conditions or simulated conditions reproduced. The undoing of all engineering is the natural reaction of nature in a non bench condition.
As stated" this problem does not occur in the south" In the south they have a hotter summer. We are focused on the hot, We are discounting the cold cycle that they do not have. While the hot cycle would cause boiling and evaporation to the fuel, one factor. The cold cycle causes the fuels water content to freeze and separate within the fuel line line causing reaction within the line. This reaction causes oxidation within the line, oxidation within the line causes a variation from the bench optimal condition.
The effects of oxidation within the fuel line can vary, however the effect of oxidation may cause more surface area to be present within the fuel line causing a reactive surface, the oxide may trap and transfer heat to the fuel or the oxide itself may be reacting with the fuel.
My first consideration would be to research the fuel line and its metal type, design, and how it has been affected by time.