Seabed gallery intakes: Investigation of the water pretreatment effectiveness of the active layer using a long-term column experiment.

Seabed gallery intake systems used for seawater reverse osmosis facilities employ the same principle of water treatment as slow sand filtration in freshwater systems. An investigation concerning the effectiveness of the active layer (top layer) in improving raw water quality was conducted by using a long-term bench-scale columns experiment. Two different media types, silica and carbonate sand, were tested in 1 m columns to evaluate the effectiveness of media type in terms of algae, bacteria, Natural Organic Matter (NOM) and Transparent Exopolymer Particles (TEP) removal over a period of 620 days. Nearly all algae in the silica sand column, 87% (σ = 0.04) of the bacteria, 59% (σ = 0.11) of the biopolymer fraction of NOM, 59% (σ = 0.16) of particulate and 32% (σ = 0.25) of colloidal TEP were removed during the last 330 days of the experiment. Total removal was observed in the carbonate sand column for algal concentration, while the bacterial removal was lower at 74% (σ = 0.08). Removal of biopolymers, particulate and colloidal TEP were higher in the carbonate column during the last 330 days with 72% (σ = 0.15), 66% (σ = 0.08) and 36% (σ = 0.12) removed for these organics respectively. Removal of these key organics through the 1 m thick column, representing the active layer, will likely reduce the rate of biofouling, reduce chemical usage and minimize operating cost in SWRO systems. The data show that the media will require several months at the beginning of operation to reach equilibrium so that high organic removal rates can be achieved. No development of a "schmutzdecke" layer occurred. The experimental results suggest that unlike freshwater slow sand filtration wherein most water treatment occurs in the upper 10 cm, in seawater systems treatment occurs throughout the full active layer depth of 1 m. The results of this study will help in designing and operating seabed gallery intake systems in varied geological conditions.

Aquifer Treatment of Sea Water to Remove Natural Organic Matter Before Desalination.

An investigation of a sea water reverse osmosis desalination facility located in western Saudi Arabia has shown that aquifer treatment of the raw sea water provides a high degree of removal of natural organic matter (NOM) that causes membrane biofouling. The aquifer is a carbonate system that has a good hydraulic connection to the sea and 14 wells are used to induce sea water movement 400 to 450 m from the sea to the wells. During aquifer transport virtually all of the algae, over 90% of the bacteria, over 90% of the biopolymer fraction of NOM, and high percentages of the humic substance, building blocks, and some of the low molecular weight fractions of NOM are removed. Between 44 and over 90% of the transparent exopolymer particles (TEP) are removed with a corresponding significant reduction in concentration of the colloidal fraction of TEP. The removal rate for TEP appears to be greater in carbonate aquifers compared to siliciclastic systems. Although the production wells range in age from 4 months to 14 years, no significant difference in the degree of water treatment provided by the aquifer was found.

Subsurface intake systems: Green choice for improving feed water quality at SWRO desalination plants, Jeddah, Saudi Arabia.

An investigation of three seawater reverse osmosis facilities located along the shoreline of the Red Sea of Saudi Arabia that use well intake systems showed that the pumping-induced flow of raw seawater through a coastal aquifer significantly improves feed water quality. A comparison between the surface seawater and the discharge from the wells shows that turbidity, algae, bacteria, total organic carbon, most fractions of natural organic matter (NOM), and particulate and colloidal transparent exopolymer particles (TEP) have significant reductions in concentration. Nearly all of the algae, up to 99% of the bacteria, between 84 and 100% of the biopolymer fraction of NOM, and a high percentage of the TEP were removed during transport. The data suggest that the flowpath length and hydraulic retention time in the aquifer play the most important roles in removal of the organic matter. Since the collective concentrations of bacteria, biopolymers, and TEP in the intake seawater play important roles in the biofouling of SWRO membranes, the observed reductions suggest that the desalination facilities that use well intakes systems will have a potentially lower fouling rate compared to open-ocean intake systems. Furthermore, well intake system intakes also reduce the need for chemical usage during complex pretreatment systems required for operation of SWRO facilities using open-ocean intakes and reduce environmental impacts.

Determination of hydraulic conductivity from grain-size distribution for different depositional environments.

Over 400 unlithified sediment samples were collected from four different depositional environments in global locations and the grain-size distribution, porosity, and hydraulic conductivity were measured using standard methods. The measured hydraulic conductivity values were then compared to values calculated using 20 different empirical equations (e.g., Hazen, Carman-Kozeny) commonly used to estimate hydraulic conductivity from grain-size distribution. It was found that most of the hydraulic conductivity values estimated from the empirical equations correlated very poorly to the measured hydraulic conductivity values with errors ranging to over 500%. To improve the empirical estimation methodology, the samples were grouped by depositional environment and subdivided into subgroups based on lithology and mud percentage. The empirical methods were then analyzed to assess which methods best estimated the measured values. Modifications of the empirical equations, including changes to special coefficients and addition of offsets, were made to produce modified equations that considerably improve the hydraulic conductivity estimates from grain size data for beach, dune, offshore marine, and river sediments. Estimated hydraulic conductivity errors were reduced to 6 to 7.1 m/day for the beach subgroups, 3.4 to 7.1 m/day for dune subgroups, and 2.2 to 11 m/day for offshore sediments subgroups. Improvements were made for river environments, but still produced high errors between 13 and 23 m/day.