Node capacitance difference between captab and AC analysis

I want to measure the effective capacitance on a particular node of my schematic ("/net1").

I tried two methods:

1. Using the captab nodetonode option in DC analysis, reading the "/net1" to "/net1" total capacitance from Capacitance Table window.

2. Adding a 1V AC source (V0) connected to the node and running AC analysis. Then get the capacitance vs frequency plot from Spectre Calculator
using the formula 1/(2*3.141593*imag(VF("/net1")/IF("/V0/PLUS"))*xval(VF("/net1"))). (This is just C = 1/(2*pi*X*f) where X is reactance and f is frequency in Hz).

I cannot understand how to interpret the results, because the AC analysis gives me a frequency-dependent capacitance, while the Capacitance Table gives a single number.

Is the Capacitance Table value supposed to give me capacitance at limit when f approaches 0 Hz from the AC analysis?

The numbers I get are totally different. The AC analysis capacitance ranges from 1 pF to 2 fF for frequency between 1 Hz and 1 THz.

But the captab value is 0.076 fF, which is smaller than all values from the plot (it's split into Variable: 0.076 fF, Fixed: 0.0).

I don't know what is the most accurate way to measure the effective capacitance on node /net1, and whether any of the results I get make sense.

Additional information:
-----------------------------
Simulator: spectre
Version icfb: 5.10.41

Thank you.