Abstract

The World Health Organization reported in 2000 that over 100 million people in Latin American and the Caribbean did not have access to proper excreta disposal facilities. Inadequate onsite sanitation treatment in Jamaica has resulted in significant nitrate and coliform contamination of groundwater. To address such issues, locally designed onsite wastewater treatment systems consisting of septic tanks and constructed wetlands had been built at two rural public schools, one in Pisgah, St. Elizabeth Parish and one in Retrieve, St. James Parish. Septic tank and subsurface flow wetland performance has been documented well for the industrialized world. To broaden the body of knowledge of such systems operating in the developing world, this report documents the performance of these septic tanks and subsurface flow wetlands operating in Jamaica, and also contains brief guidance on designing and operating these systems.Primary treatment was achieved with two-chamber septic tanks followed by horizontal subsurface flow constructed wetlands planted with local wild cane (Gynerium sagittatum) achieving secondary treatment with nutrient removal. (In Jamaica, this is referred to as tertiary treatment.) The toilet system in Pisgah had been operating for one and a half years supplied with water from a rainwater harvesting scheme. Per capita water use averaged 1.3 L/p-d and total water use averaged 264 L/d (70 USgal/d). This resulted in such low hydraulic loading that the septic tanks had a hydraulic residence time (HRT) of 29 days. The wetland never produced effluent so that it functioned as an evapotranspiration (ET) bed with an average ET rate of 0.27 USgal/ft2/d (11 mm/d). Wetland BOD mass loading averaged 0.79 kg/ha-d (0.71 lb/ac-d) and TSS entry zone mass loading averaged 0.0013 lb/ft2-d (6.4 g/m2-d). The toilet system in Retrieve had been operating for two years with a municipal water supply. Per capita water use averaged 48.3 L/p-d and total water use averaged 3,240 L/d (857 USgal/d). The average septic tank HRT was 1.2 days. Combined sewage inflow and precipitation to the wetland resulted in an average HRT of 2.2 days equivalent to a hydraulic load of 2.6 USgal/ft2-d (105 L/m2-d). Wetland BOD mass loading averaged 15 kg/ha-d (13 lb/ac-d) and TSS entry zone mass loading averaged 0.0038 lb/ft2-d (19 g/m2-d).

Five sets of water quality grab samples were collected from three points at each site. Samples from the sanitation system in Pisgah indicated an average reduction of raw sewage BOD by 78%, TSS by 85%, total nitrogen by 95%, ammonia by 99%, total phosphorus by 97%, total coliform and fecal coliform by 4 log (99.99%). Samples from the sanitation system in Retrieve indicated an average reduction of the raw sewage BOD by 50%, total nitrogen by 68%, ammonia by 97%, total phosphorus by 64%, total coliform by more than 3 log and fecal coliform by more than 4 log. An increase in TSS for the system at Retrieve may have been caused by the effluent sampling method. Average nitrate levels were below 1 mg/l throughout both sanitation systems