Dr. Dayananda Vithanage, P.E.
Dr. Patrick K. Sullivan, P.E.

Oceanit Laboratories, Inc.
Honolulu, HI



Municipal wastewater has density characteristics similar to fresh water. Therefore, in a homogeneous ocean the waste field will always rise to the surface. However, under adequately stratified conditions, effluent surfacing is suppressed. Variability of water density and velocity profiles in relatively shallow depths strongly influence wastewater discharge outfall design. Degree of density stratification and ambient currents are major factors that determine the extent of wastewater dilution during the buoyancy phase (initial or near-field dilution) as well as the equilibrium depth of the waste field.

In order to investigate variability of parameters due to internal wave effects in the vicinity of steep slopes and their influence on mixing and dispersion, an experiment funded by the Office of Naval Research was conducted adjacent to an ocean wastewater outfall in Mamala Bay, Oahu, Hawaii. The site is located on the south shore of Oahu approximately 2 miles offshore at a water depth of 230 feet. Wastewater is released at an average rate of 75 million gallons per day (mgd) through a diffuser approximately 1400 feet long. The dispersion characteristics and the fate of the wastewater after discharge is of great importance because of extensive coral reefs in the nearshore area and relative proximity of the outfall to popular recreational beaches. The study location is shown in Figure 1.


A thermistor string with 14 thermistors at 10-foot intervals and a downward-looking Acoustic Doppler Current Profiler (ADCP) were moored at a site close to the outfall diffuser at a depth of 200 feet. The measurement profile was designed to sample both temperature and velocity from 50 feet below the surface to the bottom to avoid navigational hazards. Velocity and temperature profiles were measured at one-minute intervals. Data were downloaded monthly from loggers; conductivity and temperature profiles were measured monthly at the site using a Seabird CTD profiler to supplement data and to check for possible drift in thermistor measurements.


Temperature and velocity data were used to investigate wastewater dilution and plume equilibrium depths. In addition, statistics for all data sets were calculated. In order to investigate high frequency variability, one-hour segments of data were analyzed for variance. Hourly calculated variances at all depths were averaged. Average temperature variance distribution for September 1997 are shown in Figure 3.

The data was high passed to filter out low frequency phenomena that dominated the raw data. A Fourier filter with a cut off at 60 minutes was used for this purpose. A cosine taper was used at the cut off to reduce data contamination. The high passed temperature data had a weak periodicity at about 40 minutes. High passed data was demodulated at a period of 40 minutes to remove noise. The original data, high passed data and demodulated high passed data are shown in Figure 2. Spectral density of high passed data at different depths was calculated using a data length of 256. The spectral density distribution of high frequency temperature data over the depth for the month of September 1997 is shown in Figure 4.


1. Raw temperature data shows sudden changes at approximately tidal periods. High passed data shows high perturbations when this occurs. These perturbations show a periodicity of about 40 minutes and die off gradually, until excited again by a sudden temperature change. Could the perturbations be due to solitons associated with tides? If these are solitons, are they produced by the interaction of the tide with the island flank?

2. High passed data show high energy at about 0.0234 cycles per minute (43 minutes). The phenomena are intermittent and there are periods of no activity in between. Is this related to the amplitude of the surface tide?

3. Similar high energy density is seen at the same frequency for all components of velocity. Could this be due to resonance of an internal wave with the bottom slope?







Figure 2. Time series of raw data, high passed data, demodulated high pass data (0.0234cpm)
for temperature at 110 feet depth, and surface tide.

Figure 3. Average hourly variance of Temperature for September 1997.

Figure 4. High frequency spectral density distribution over depth.