Effect of pond depth on removal of nitrogen in duckweed stabilization ponds

The effect of pond depth on nitrogen removal in duckweed stabilization ponds was studied in a pilot plant consisting of two lines with seven duckweed ponds in series, with different depths and fed with effluent of a laboratory scale UASB reactor. Three experimental conditions were studied: l) Pond depth 0.7 m, HRT = 21 days, 2) Pond depth 0.4 m, HRT = 12 days, 3) Pond depth 0.4 m, HRT = 21 days. The systems were evaluated based on pH, temperature and oxygen profiles, organic matter removal (BOD 5 ), nitrogen transformations, biomass production and biomass nitrogen content. Average total nitrogen removal rates were 598 mg N m -2 d -1 for DSP 1, 589 mg Nm -2 d -1 for DSP 2 and 482 mg N m -2 d -1 for DSP 3. In spite of the lower nitrogen removal rate in DSP 3, it has the higher removal efficiency (44 %, 43 % and 62 % for DSP 1, 2 and 3 respectively) due to the lower surface loading rate in this system. This shows that using the percentage of removal as a parameter for comparison should be done with care and the operational parameters of the compared systems should be taken into account. Denitrification was the most important nitrogen removal mechanism for the three DSPs. Nitrogen removal via biomass production was the second most important removal mechanism for the three experiments. Pond depth does not seem to determine nitrification or denitrification,. Nitrification seems to be related to surface organic loading rate, while denitrification was related to BOD availability. The comparison between two pond systems with different depths, but operated at the same hydraulic surface loading rate (DSP 1 and 2) showed similar nitrogen removals in the shallower system as in the deeper system. This suggests that duckweed pond system could be designed with the shallower depth without affecting surface loading and nitrogen removal efficiency. Nitrogen removal appeared to be governed by the surface loading rate rather than by the hydraulic retention time.