Harmonic AnalysisΒΆ
The equation that defines the water level changes due to tide is:
![\widehat{h} = H_0 + \sum_{i=1}^n {\widehat{f_i} H_i \cos \left ( a_i \widehat{t} - \left ( \kappa_i - \left [ V_o + \mu \right ]_i \right ) \right )}\\](../../_images/math/4c6afa81fe29ccca7f4882c4a024797e3daafa37.png)
![where:\\
\widehat{h} &=\mbox{measured water level (1 dimensional array)}\\
H_0 &=\mbox{average water level}\\
n &=\mbox{number of constituents}\\
\widehat{f_i} &=\mbox{nodal factor for constituent }i\mbox{ (1 dimensional array)}\\
H_i &=\mbox{amplitude of constituent }i\\
a_i &=\mbox{speed of constituent }i\\
\widehat{t} &=\mbox{time (1 dimensional array)}\\
\kappa_i &=\mbox{phase angle of constituent }i\\
\left [ V_o + \mu \right ] &=\mbox{equilibrium argument for constituent }i](../../_images/math/47c12fcc24ad16d59f747b341a26f2559353bb60.png)
Typical tidal analysis programs specify the node factor at the center of the analyzed time-series. I think this is something that is true for a 29-day series, but for longer analysis, the node factor at each measurement should be calculated. This is what TAPPY does.