I apologize for long reply of how the watch was made rather than just saying what the current time is.
(Short read)
Its only a paper reduction in power and they are replacing the tall tower due to its poor condition.
(Long in-the-weeds explanation)
680 was always limited in field because the height of the day tower was over the maximum for that class of operation. Up until the mid 1980's, power levels were set at defined levels such as 250,500, 1,000, 5,000, 10,000, 25,000, 50,000 watts respectively. If your tower was too efficient and exceeded the RMS field intensity for that class of operation, the station had to limit that field by placing a series resistor in-line with the antenna. In 680's case up until the FCC moved away from set levels, they most likely used a dropping resister to absorb 2,000 watts. At some point in time years ago, they decided to add the tall tower to support the FM and use it for daytime and gain some control of it by adding it to the night time array as what is described as a negative tower. In these cases the tall tower is very efferent but the lowest contributor to the distant field RMS. In this power flow scheme, the tall tower receives its excitation from the other elements in the array and that is power returned in-phase to the phasor common point to make up for the loss in field. Today, with the non-standard power levels, the resister is not needed and you can simply license the thing for 8,000 watts. With the night directional situation, the tall tower was likely a minor contributor to pattern shape given its low field contribution, but an array stability issue. With today's RF molding capabilities utilizing electromagnetic incremental analysis, it is relatively easy to factor out the unused tower and perhaps de-tune it to where it is invisible during directional operation. Feeding negative towers back into the phasor is also a detriment to a linear transmitter load impedance bandwidth.
Best regards.
w/