Substrate clogging is by far the biggest operational problem of vertical flow constructed wetlands. The term "substrate clogging" summarises several processes which lead to reduction of the infiltration capacity at the substrate surface. The lower infiltration rate causes a reduced oxygen supply and further leads to a rapid failure of the treatment performance. Reasons for substrate clogging include accumulation of suspended solids, surplus sludge production, chemical precipitation and deposition in the pores, growth of plant-rhizomes and roots, generation of gas and compaction of the clogging layer. However, it is not clear how much each process contributes to the clogging process. Detailed investigations were carried out at pilot-scale constructed wetlands (PSCWs) using a variety of methods: e.g. soil physical investigations, microbial methods, and various analysis methods of drinking water and wastewater. The paper shows the results of these investigations and presents an equation to calculate the theoretical clogging time.

Surface water pollution is one of the serious environmental problems in urban centers in Nepal due to the discharge of untreated wastewater into the river-system, turning them into open sewers. Wastewater treatment plants are almost non-existent in the country except for a few in the Kathmandu Valley and even these are not functioning well. Successful implementation of a few constructed wetland systems within the past three years has attracted attention to this promising technology. A two-staged subsurface flow constructed wetland for hospital wastewater treatment and constructed wetlands for treatment of greywater and septage is now becoming a demonstration site of constructed wetland systems in Nepal. Beside these systems, five constructed wetlands have already been designed and some are under construction for the treatment of leachate and septage in Pokhara municipality, wastewater in Kathmandu University, two hospitals and a school. This paper discusses the present condition and treatment performance of constructed wetlands that are now in operation. Furthermore, the concept of the treatment wetlands under construction is also described here. With the present experience, several recommendations are pointed out for the promotion of this technology in the developing countries.

To treat the wastewater of a hospital at Dhulikhel/Nepal, a two-stage constructed wetland was built with a settlement tank, a horizontal flow bed as first stage and a vertical flow bed as second stage. The plant is operated without electric power. The aim was the elimination of organic compounds, nitrification and a significant reduction of indicator bacteria. Different phases of operation (high and low water level within the soil profile, serial operation, one stage operation) were investigated, of which the serial operation with high water level in the horizontal flow bed and low water level in the vertical flow bed showed the best elimination performance. The areal removal rate constants (k-values) turned out to be very high (especially of the vertical flow bed) compared with literature values of other subsurface flow constructed wetlands. For the vertical flow bed kCOD was 0.22 m/d and kNH4-N was 0.85 m/d during serial operation. For kNH4-N a strong correlation with the hydraulic loading rate and the COD inlet concentration was found.

Since 1991 the institute for water provision has run two vertical flow constructed wetlands for domestic wastewater treatment at two farm houses (8 p.e.) in Upper Austria. The systems are designed for elimination of organic compounds and for nitrification which was no problem even under winter conditions. In 1995 two methods were tested to achieve denitrification also in both systems. System A is a one-stage system, system B has two stages operated in series. The approach to the one-stage system consisted in pumping a part of the nitrified effluent from the soil filter back to the settling tank of the influent, where the raw wastewater mixes with the nitrified effluent-water. The approach to the two-stage system B consisted in adding an external carbon source (methanol) into the second, water saturated stage. The research was undertaken during several experimental stages to test the influence of the recirculation ratio (system A) respectively of the feeding (system B; intermittent, continuous, batch). The highest elimination rates could be achieved with system B (dosage of methanol) during experimental stage 1 (intermittent loading four times a day). The mean elimination rates were 82% for Ninorg and 78% for TN. The TN elimination performance of system A was only a little lower (72%). The elimination of COD, BOD5 and TOC was no problem throughout the whole experimental period (effluent concentrations of both systems were well below the Austrian standards).

Constructed Wetland (CW) technology is still not wide spread in developing countries despite having great potential. This paper describes an approach carried out in Nepal to transfer CW technology for wastewater treatment. Three CWs (hospital wastewater treatment--20 m3/d, greywater treatment of a single household, septage treatment--40 m3/d) were built and two have been investigated so far. All systems are subsurface flow systems with at least one vertical flow stage. Their treatment efficiency turned out to be very high. Median load elimination rates of the hospital wastewater and greywater treatment plants were for TSS: 97 to 99%; BOD5: 97 to 99%; COD: 94 to 97%; NH4-N 80 to 99%; PO4-P: 5 to 69%; Total Coliforms: 99.87 to 99.999% (3-5 log steps). Beside the treatment task the plants play an important role as demonstration sites to make common people and especially decision makers aware of the existing environmental problems and one possible solution. Several recommendations are pointed out to promote the technology in developing countries.