Island Surge and Wave Measurements and Modeling
       Home    

Guam - Data Considerations and Analysis Methods

The reef is wide, flat, shallow, and highly dissipative, resulting in infragravity wave (IG) dominated spectra during storms. The periodogram in Figure 1 illustrates the greatest energy at typical IG frequencies around 0.05Hz. Figure 2 shows a 17-minute record of the shoreward most Sea-Bird (SB-1) pressure sensor (top frame) and the same signal after a 20s bandpass filter was applied (middle frame) that removed most of the incident waves, leaving an infragravity wave signal. The lower frame compares the seaward most Sea-Bird sensor (SB-4) with SB-1. It appears that the shoreward gauge has longer period IG waves than the gauge closer to the reef break.

Sea-Bird wave records were kept to 17-minutes due to limited memory, battery life, and requirement to operate at least 6 months without servicing. Infragravity waves observed during the ManYi storm had periods between one and three minutes, which resulted in too few waves in the 17-minute record to use spectral analysis to reliably determine wave height and period. Vendor software did high and low frequency spectra truncation, as well as band averaging, which would not resolve the infragravity wave band. That also influenced the significant height determination. We decided to compute significant height in the time-domain using four standard deviations of the 17-minute pressure time series. Since infragravity waves dominated in the storms and were shallow-water waves, pressure attenuation corrections of the wave signal were not necessary. The incident wind-waves were likely shallow-water waves as well since the depths were about one meter.


Figure 1

Figure 2
Figure 3 also illustrates that care must be taken when making mean water level measurements from short time averages when infragravity waves are present. The top frame in this figure is a 17 minute record of the pressure signal from SB-1 sampled at 2Hz. The lower frame has a 20s and 300s band-averaged filter applied. The 300s filtered signal shows the variability that would be observed in a 5-minute tide estimate made over this this 17 minute record - about 9 cm difference depending when the average was made.

An infragravity wave field can have complex alongshore and cross-shore elevation structure over a reef. There can be a mix of leaky and edge waves with many modes possible. A cross-shore array of four pressure sensors can not resolve the details of an infragravity wave field. The goal of this research, nevertheless, was to measure extreme water levels during storms and the sensors do provide that data.


Figure 3 (Elapsed Time, s)

Normalized Water level

Mean water levels were made relative to SB-3. SB-3 was selected as the reference elevation since it appeared to have a more stable mean than SB-4, which was more influenced by incident waves. The gauges were not surveyed in at installation but are level adjusted to SB-3 based on comparisons of mean water level comparisons when water levels and wave heights were low. For SB-1, SB-2 and SB-3 this was when the mean level was greater than 0.1 m and less than 0.3m and wave height was less than 0.2 m. SB-4 limits were when mean levels were greater than 0.1 m and less than 0.4m and wave height were less than 0.4 m (Figure 4; top and middle frames). Water level differences at synchronous times were made with SB-3 when these conditions were met. Monthly averages of these differences were made over the two year period (Aug 2005 to Jul 2007) and were found to be consistent, +/- ~1cm for SB-1 and SB-2, +/- 4cm for SB-4. The same pressure offset was computed for the Aquadopp sensors relative to SB-3. The co-located Aquadopp and Sea-Bird sensors provided verification water level at each location (Figure 4) with differences generally within 1-2 cm.

Figure 4

Figure 5
An intercomparison of co-located gauges reveal excellent results during the first deployment (27 June - 18 Feb). Figure 5 is a water level comparison of Aquadopps and Sea-Bird gauges from the first storm (7-8 Jul 2006) after the initial Aquadopp deployment (27 Jun 2006). Water level comparison of cross-reef gauges indicate little difference between locations 1, 2, and 3, but do show setup from location 4 up to 20 cm. The Aquadopps and Sea-Birds both measured approximately the same setup. After the second Aquadopp deployment (18 Feb 2007) the pressure signals on the Aquadopp gauges degraded, possibly due to an obstruction in the orifice. By July 2007 when the Man-Yi storm occurred the pressure signal of gauges #1 and #3 only responded to changes of longer period than infragravity waves (e.g., tides) (Figure 5). The second deployment (18 Feb - ?) revealed problems with the aquadopp pressure. Ports which we suspect were caused by the orifices being partially impacted by antifouling grease. Aquadopp #2 had a sawtooth response to infragravity periods so it made a poor estimate of wave height (4-std). Aquadopp #4 appeared to work but with frequent spikes in the signal. Sea-Bird gauges should be used for water level and wave height after 18 Feb 2007 and Aquadopps used for current measurements.

Figure 6