WaterRenew : wastewater polishing using renewable energy crops

The research described in this thesis is part of a wider EU-LIFE project, the WaterRenew project. The WaterRenew concept can be described as the recovery of nutrients from wastewater which can lead to eutrophication of surface waters, by irrigation of short rotation coppice in order to fertilise them. Such systems have been proven to function properly as nutrient removal systems when studied for N removal and have already been successfully and commercially implemented in different countries. However, the factors potentially preventing them from operating sustainably have not been identified nor their upper limits quantified with confidence. A WaterRenew system can indeed be looked at as a unit composed with three main compartments; soil, soil water and plant. Therefore, the sustainability of such a system will be compromised if at least one of these compartments is changed irreversibly. The limits can be hydrological with constant runoff or drainage being induced. They can also be chemical with inadequate amounts of nutrients removed from the effluent applied or irreversible accumulation of nutrients in soil. Finally, these limits can be physiological with the trees’ health being irreversibly compromised. Moreover, the relevance and effectiveness of such a system under UK conditions has not been established yet. In this context, a field trial was set up at Cranfield University sewage treatment works where the secondary treated effluent was irrigated on to Salix viminalis, Populus trichocarpa and Eucalyptus gunnii trees planted at a density of 13,060 trees.ha-1, on a chalky clayey soil, in order to maintain soil water content at field capacity. To tackle more specifically P fate processes understanding, an independent P leaching soil column experiment was also set up. With the latter settings, it was possible to apply high volumes of effluent (3625 mm for willow, 2895 mm for poplar and 3345 mm for eucalyptus for the 2 years of irrigation) and high amounts of nutrients (1023 kg-N.ha-1 and 134 kg-P.ha-1 for willow, 834 kg- N.ha-1 and 108 kg-P.ha-1 for poplar and 946 kg-N.ha-1 and 127 kg-P.ha-1for eucalyptus for the 2 years of irrigation). It was found that irrigation with effluent increased significantly tree yields so that they were within the range reported in the literature for willow and eucalyptus but slightly lower for poplar. The trees uptook between 20 % and 50 % of the total amounts of N and P applied with eucalyptus uptaking more nutrients than willow, which in turn took up more than poplar. Then, it was found that irrigation did not have any significant effect on N and P in soil and the amounts applied remained very low compared to the existing nutrients soil pools. However, irrigation did have a significant effect on increasing K and Na in soil. Na increased enough to induce a significant increase in soil SAR but soil remained neither saline nor alkali. The trees had a smaller impact on soil chemistry. Finally, it was found also that irrigation did not have any significant effect on N and P in soil water with no P detectable in any of the soil water and groundwater samples during the whole experiment. Irrigation did, however, increase significantly K and Na concentrations in soil water and for K also in groundwater. From the point of view of nutrients removal, although a tree effect was measurable, it was not as important as the functions of the soil. Thus, when a WaterRenew system is maintained under a hydrological constraint, with the soil moisture kept at field capacity, it was still possible to apply high volumes of effluent, even on a clayey soil. In addition, the consequent high amounts of nutrients applied were efficiently retained between tree uptakes and mainly soil organic and inorganic nutrients’ pools. Indeed, the amounts of nutrients lost by drainage remained low (<10 % of the total amounts applied) for N and P and groundwater was efficiently protected from pollution on this site. On P dynamic processes understanding, it was found that P leaching patterns depend highly on soil moisture and to a lesser extent on the amount of P applied. When soil is saturated, P will start leaching even when applied at a very low concentration. A model, the WR_MODEL, was developed which integrates the observations, measurements and understanding of Cranfield University sewage treatment work field trial and soil column experiment into a model. The purpose of WR_model is to help the design and implementation of a WaterRenew system in any location as long as climatic and soil data are available. The model default values are for England and Wales climatic and soil data.