On fitting the k-C* first order model to batch loaded sub-surface treatment wetlands.

The k-C* first order model was fit to time-series COD data collected from batch-loaded model wetlands. Four replicates of four plant species treatments; Carex utriculata (sedge), Schoenoplectus acutus (bulrush), Typha latifolia (cattail) and unplanted controls were compared. Temperature was varied by 4 degrees C from 24 degrees C to 4 degrees C to 24 degrees C over a year-long period. One mathematical fit was made for each wetland replicate at each temperature setting (192 fits). Temperature effects on both parameters were subsequently estimated by fitting the Arrhenius relationship to the estimated coefficients. Inherent interactions between k and C* make values dependent on sample timing and statistical technique for either time series (batch load) or distance profile (plug flow) data. Coefficients calibrated using the Levenberg-Marquardt method are compared to values previously reported using a nonlinear mixed effect regression technique. Overall conclusions are similar across approaches: (a) the magnitude of the coefficients varies strongly by species; (b) the rate constant k decreases with increasing temperature; and (c) temperature and species variation in the residual concentration C* is greater than the variation in k, such that variation in k alone is a poor predictor of performance. However, the magnitudes of the coefficients, especially the rate parameter k, vary between the statistical techniques, highlighting the need to better document the statistical routines used to calibrate the k-C* model.

Does batch operation enhance oxidation in subsurface constructed wetlands?

Two side-by-side experimental sub-surface flow systems allowed direct comparison of wetland performance under batch and continuous-flow operation. One system consisted of microcosm "columns" operated in 20-day batch mode while the second consisted of continuous-flow "cells" operated at a five-day residence time. Both systems treated identical synthetic domestic wastewater for two years and then treated identical synthetic mine-impacted water for one year. Each system had replicates planted with Typha latifolia, Scirpus acutus and unplanted controls. Temperature was cycled annually between 4 to 24 degrees C. Results indicated that plant species, season, and mode of operation interacted strongly in controlling dynamics of COD, nitrogen species, phosphate, sulfate, and redox potentials. In batch-loaded columns, between-species differences in oxidation and COD removal were large in winter, during plant dormancy, but absent in summer; COD removal, sulfate concentration, and redox potentials were closely correlated, suggesting that variation in root-zone oxygenation due to seasonal plant growth patterns and temperature-dependent plant and microbial respiration may explain observed differences. In the continuous-flow cells, species and seasonal differences were minimal or non-existent, indicating that under continuous-flow operation plants either did not influence root zone oxidation or that this influence had no effect on wetland performance for COD and nutrient removal or sulfate reduction.