Nutrient removal from agricultural run-off in demonstrative full scale tubular photobioreactors for microalgae growth

NUTRIENT REMOVAL FROM AGRICULTURAL RUN-OFF IN DEMONSTRATIVE FULL SCALE TUBULAR PHOTOBIOREACTORS FOR MICROALGAE GROWTH Joan García1*, Antonio Ortiz1, Eduardo Álvarez1, Vojtech Belohlav1,2, María Jesús GarcíaGalán1, Rubén Díez-Montero1, Juan Antonio Álvarez3 and Enrica Uggetti1 1GEMMA-Environmental Engineering and Microbiology Group, Department of Civil and Environmental Engineering, Universitat Politècnica de Catalunya-BarcelonaTech, c/ Jordi Girona 1-3, Building D1, E-08034 Barcelona, Spain 2Department of Process Engineering, Czech Technical University in Prague, Technicka 4, 166 07 Prague, Czech Republic 3AIMEN Technology Centre, c/ Relva 27 A, Torneiros, E-36410 Porriño, Pontevedra, Spain *Corresponding author: Tel.: +34 934016464 Fax. +34 934017357 E-mail address: [email protected] Abstract The objective of this paper is to present the design, construction and operation (during one year) of 3 full scale semi-closed, horizontal tubular photobioreactors (PBR, 11.7 m3 of volume each) used to remove nutrients of a mixture of agricultural run-off (90%) and treated domestic wastewater (10%). PBRs were located outdoor and have 2 paddlewheels (engines of 0.25 kW) to ensure the movement of the mixed liquor. The microalgal biomass produced in the PBRs was harvested in a static lamella settling tank in which a polyaluminium chloride coagulant is applied. Each PBR treated in average 2.3 m3/d, being the actual mean hydraulic retention time 5 d. PBRs were submitted to strong seasonal changes regarding solar radiation and temperature, which had a direct impact in the activity of microalgae and the efficiency of the system. Higher mixed liquor pH values were registered in summer (daily average > 10). These high values were not observed in the effluents because the system was designed to discharge the mixed liquor (effluent) only at the end of night, when pH reached the lowest daily values (around 8.5). Most of the influent and effluent nitrogen content was inorganic (average of 9.0 mg N/L and 3.17 mg N/L, respectively), and in the form of nitrate (62% and 50%, respectively). Average nitrogen removal efficiency was 65%, with values of around 90% in summer, 80% in autumn, 50 % in winter and 60% in spring. Most of the influent and effluent phosphorus content was in the form of ortophosphate. Influent average was 0.62 mg P/L, but with great variations and in a considerable number of samples not detected. Removal efficiency (when influent values were detected) was very high during all the study, usually greater than 95%, and there were not clear seasonal trends for efficiency as observed for TIN. Volumetric biomass production greatly changed between seasons with much lower values in winter (7 g VSS (volatile suspended solids)/m3·d) than in summer (43 g VSS/m3·d). Biomass separation efficiency of the settler was very good in either terms of turbidity and total suspended solids, being most of the time lower than 5 UNT and 15 mg/L, respectively. Overall this study demonstrated the reliable and good effectiveness of microalgae based technologies such as the PBR to remove nutrients at a full scale size. Keywords: phytoremediation, cyanobacteria, agricultural drainage, eutrophication, high rate algal ponds 1. Introduction Changes in the nutrient biochemical flows due to anthropogenic activities are one of the main environmental challenges that humanity must face in the coming decades. The alteration of the cycles of nitrogen and phosphorus (N and P) is already considered of high global risk, with unfavourable effects leading to unknown impacts (Steffen et al., 2015). Urban and agricultural discharges of contaminated or insufficiently treated water are the main cause for the imbalance of these biochemical cycles. Nowadays, most of the aquatic ecosystems are receiving these nutrient enriched discharges, being their eutrophication an unequivocal signal of it. Globally, more than 450 coastal areas are affected by severe eutrophication (Selman et al., 2008). Ecological engineering techniques can be used to reverse this contamination situation in many cases, allowing also for the restoration of these aquatic ecosystems. In particular, treatment wetlands have been extensively used in recent decades as effective systems for the treatment of urban, agricultural and even industrial wastewater; a vast array of literature with hundreds of examples at full scale is available (Ávila et al., 2013; García et al., 2010). There is much less experience with other types of ecological engineering techniques such as microalgae systems, despite the fact that microalgae based wastewater treatment systems were developed more than 50 years ago, specifically to treat urban wastewaters (García et al., 2006). Therefore, it is necessary to show and demonstrate the potential of these microalgae technologies at full-scale. One of the most powerful advantage of microalgae systems in comparison to other technologies is that harvested microalgae biomass can easily be valorised as a bioproduct and/or energy, which is extremely interesting within the framework of the circular economy. Up to date, most of the studies devoted to phytoremediation of agriculturalrelated wastes by means of microalgae have focused on lab-scale experiments to treat industrial effluents, such as those from dairy farms (Labbé et al., 2017), palm oil mills (Kamyab et al., 2015) or rice mills (Kumar et al., 2016). The treatment of aquaculture effluents and diluted pig slurry treatments were also investigated in different works (Ansari et al., 2017; Lananan et al., 2014; Ledda et al., 2016). The capacity of microalgae to remove pesticides from agriculture run-off was also evaluated by Matamoros et al. (2016). In all cases, however, only lab-scale experiments were performed, usually with microalgae cultures grown on synthetic media and aseptic conditions. To the authors’ knowledge, only two recent studies have evaluated the feasibility of integrating agricultural run-off treatment and biomass production at real scale. Bohutskyi et al., (2016) studied the phytoremediation of agricultural run-off by filamentous green microalgae (Cladophora sp. and Rhizoclonium sp.) in an Algal Turf Scrubber (ATS®), treating 10 million gallons per day. The authors obtained a maximum monthly productivity of 22 g/m2·d (measured as volatile suspended solids) and a suitable feedstock to obtain biogas after anaerobic digestion. Furthermore, diluted digestate from anaerobic digestion was used as nutrients supplement to cultivate more valuable microalgae species. The second study by García-Galán et al. (2018) evaluated the efficiency of a large-scale photobioreactor treating agriculture run-off and also obtaining microalgae biomass as added-value product. Results showed a maximum biomass production of 76.4 g/m3·d (measured as total suspended solids) in April, and a total N elimination ranging from 84% to 95%. In the present paper we describe the experience gained on the design, construction and operation during the first year after the start-up (from May 2017 to May 2018) of 3 full scale photobioreactors (PBRs) fed with a mixture of agricultural run-off and treated domestic wastewater. The microalgae biomass produced in the photobioreactors was harvested in a static lamella settling tank. All these units were constructed in the framework of the innovation European project INCOVER (http://incover-project.eu/). These PBRs are part of a complex installation aiming to efficiently treat wastewater and produce bioenergy, bioproducts and reclaimed water for irrigation. A brief description of the entire experimental site can be found in Uggetti et al., (2018). This study is exclusively focused on the PBRs functioning and their auxiliary elements. The INCOVER project will be operative till May 2019. 2. Materials and Methods 2.1. Photobioreactors design The PBRs are located in the Agròpolis experimental campus of the Universitat Politècnica de Catalunya-BarcelonaTech (UPC) (41.288 N, and 2.043 E UTM), very near to Barcelona’s airport (Figure 1 and Figure A1 in the Appendixes). The PBRs and their auxiliary elements were conceived, designed and constructed by the GEMMA Research Group of the UPC in collaboration with the company Disoltech S.L after several previous investigations (García-Galán et al., 2018; Solimeno et al., 2017; Uggetti et al., 2018). These PBRs are tubular horizontal semi-closed reactors, each consisting of 2 lateral open tanks made from 10 mm polypropylene (5 m width, 1 m length and 0.6 m height, nominal volume of 1.25 m3 each at design water depth). Both tanks are connected through 16 low density polyethylene tubes (0.3 mm thick, 125 mm diameter and 47 m length, nominal volume of 9.2 m3 for all tubes together) (Figure 2 and Figure S2). These tubes lie down on a plastic covering sheet in order to ensure separation from the ground, and they are protected by agricultural anti-birds nets. The total useful volume of each PBR is 11.7 m3 (approximately 20% corresponding to the tanks, and 80% to the tubes). In each open tank, a paddlewheel with eight blades (1 m width, 0.35 m long) is installed 1.8 m away from the external edge and at 3 cm height from the bottom. An engine (0.25 kW) connected to each paddlewheel provides a turning speed which can be changed from 0 to 12 rpm. Rotation of the paddlewheel makes the mixed liquor contained in the tank move from a shallow water sector to a deep one. Difference in pressure head causes a gravity flow through 8 tubes from the deep side of one tank to the shallow side of the opposite one. Then again, the flow is moved by the paddlewheels to the deeper side part of the tank, and then it returns to the shallow side of the first tank through the other 8 tubes, and so on (Figure S2). Each tank has an inclined dam in the deep sector, which assists in maintaining the two different surface water levels and avoids big waves within the tank (Figure 2). Both open tanks ensure and favour the homogenous distribution and mixing of the liquor and also the release of the exceeding dissolved oxygen accumulated along the closed tubes. Figure 1. Location map of the facilities (left) and plan view of unit processes in INCOVER project site (right). Figure 2. Picture of the 3 photobioreactors located in parallel showing the tubes and the open tanks (upper) and lateral schematic representation of one tank (lower). Note the paddlewheel with the eight blades (which has a cover), and the inclined dam. Each PBR is equipped with online sensors of pH (Hatch Lange SL., Spain), dissolved oxygen (DO) (Neurtek, Spain) and temperature (Campbell Scientific Inc., USA) in one of the two open tanks. Data of these parameters are taken every 5 s and recorded and stored each 60 s in a datalogger (Campbell Scientific Inc., USA). PBRs also include a water level sensor (Wras, UK) to control filling and emptying operations. They also have an automatic CO2 injection system (tubing, valves and pressure sensor), but at the time of the present work it was not being used. The three PBRs were installed in winter 2016-17, and were inoculated at the end of April 2017 with a mixed culture grown in experimental high rate algal ponds fed with urban wastewater (Gutiérrez et al., 2016). A volume of 10 L was added to each PBR, with a volatile suspended solids (VSS) concentration of approximately 220 mg/L. The inoculum consisted of a community of bacteria, microalgae, protozoa and small metazoa, but mostly dominated by green microalgae Chlorella sp. and Stigeoclonium sp., and diatoms Nitzschia sp. and Navicula sp. (Gutiérrez et al., 2016). Note that Stigeoclonium is a branched filamentous microalgae which in natural aquatic environments usually grows attached to submerged surfaces. In the particular case of the inoculum used here, it was growing in the form of flocs submerged in the mixed liquor of the high rate algal ponds. After inoculation, the three PBRs were operating in parallel and fed with a mixture of agricultural run-off and domestic wastewater (design ratio of 6:1 respectively, although the actual ratio was slightly higher). Total design flow was 7 m3/d (6 m3/d of agricultural run-off and 1 m3/d of treated domestic wastewater) (see Supplementary Materials, Methods section). 2.2. Auxiliary elements description and system operation Treated domestic effluent is obtained from an aerated septic tank which receives the wastewater of the main building of the campus Agròpolis (~20 persons, without overnight stay), whereas agricultural wastewater comes from a drainage collection channel (see Figure 1). Figure 3 shows a process flow diagram of the PBR and their auxiliary elements. Daily operation cycle starts at 4:30 AM when the treated domestic wastewater, stored in a cylindrical glass fiber tank (TK-103, 1 m3) discharges in a cylindrical polyethylene homogenization tank (HT-102, 10 m3, provided with a sampling port) through stream line 3. This operation is done by means of a centrifugal pump P104 (14.4 m3/h) during a maximum time of 30 min. Treated domestic wastewater continuously reaches TK-103 by the stream line 2, that conveys treated wastewater to the tank thanks to a submersible pump located in the aerated septic tank. When more than 1 m3 domestic wastewater is produced per day, TK-103 remains full and the remaining wastewater gets out through a weir, reaching a general by-pass. Note that the amount of treated domestic wastewater is often lower than 1 m3 (i.e., during the weekends). Figure 3. Process flow diagram of the photobioreactors and their auxiliary elements. At 5:00 AM, 6.0 m3 of agricultural run-off are pumped by a submersible pump (P101), from the nearby channel to the homogenization tank during a period of approximately 3 hours (stream 1). Agricultural run-off and domestic wastewater are mixed in that tank and discharged in a unique outflow at 8:00 AM (streams 4a, 4b and 4c) to the 3 PBRs by means of 3 centrifugal pumps (P-201, P-202 and P-203, time and level (LS) controlled) during a period ranging from 1 to 1.5 h. The homogenization tank has an internal recirculation pump to ensure the complete stirring of the tank. Note that stream lines 1 and 3 have flowmeters (FM) installed (Siemens, Germany). The homogenization tank has the necessary tubing and valves for maintenance and discharge to the general by-pass. At 7:00 AM, and before filling the PBRs with fresh wastewater, a designed volume of 2.3 m3 of mixed liquor of each PBR are simultaneously discharged to 3 circular glass fiber outflow storage tanks (TK-301, TK-302 and TK-303, 2.7 m3 each) through stream lines 5a, 5b and 5c. This operation is done by 3 centrifugal pumps (P-204, P-205 and P206) which provide a constant flow of approximately 3 m3/h. Therefore, that volume is evacuated from the PBRs daily in 45 min approximately. Each outflow storage tank has two different water level sensors (LS, located at the top and at the bottom) that control the filling and emptying pumps, as well as internal submerged pumps for complete stirring of the tanks and different sampling ports. As a result of the operation procedure described above, PBR functioning is done in parallel and in semi continuous mode. Note that the emptying and filling processes are done early in the morning (once a day) in order to promote biomass growth when sunlight is available. PBR theoretical hydraulic retention time (HRT) is 5.0 d. PBRs are usually operated with a paddlewheel speed ranging from 10 to 12 rpm, providing a depth in the shallow water sector of 0.25 m, and 0.45 m into the deep sector (Figure 2). This 0.2 m difference in pressure head gives a theoretical hydraulic speed of 0.25 m/s inside the tubes, which ensures turbulent flow (Reynolds number, Re >30,000 at 20 °C ). HRT within the tubes at design speed is 3.25 min, and 0.81 min in the two tanks (both together). A single drop of water theoretically gives daily 360 loops inside the PBR, and the paddlewheels move more than 4,200 m3 of water every day. PBRs have the necessary tubing and valves installed for maintenance and discharge to the general bypass. TIN was therefore successfully removed with an annually global average of 65%, but removal efficiencies changed during the different seasons due to variations in environmental conditions. Efficiencies decreased from summer (91%) to autumn (83%), winter (49%) and spring (59%). Lower solar radiation and temperature in winter resulted into lower microalgae activity and growth, and higher concentrations of TIN in PBR mixed liquor (Figure 7). Also an atypical cold and rainy weather in a great part of spring gave place to not low concentrations in TIN PBR mixed liquor. García-Galán et al., (2018) evaluated the efficiency of a similar PBR treating the same water in spring (mostly in April) and obtained a TIN removal ranging from 84 to 95%. Time, months May Jun Jul Sep Oct Nov Dec Jan Feb Mar Apr May TIN, mg/L 0 5 10 15 20 Influent Effluent Time, months May Jun Jul Sep Oct Nov Dec Jan Feb Mar Apr May PO43--P, mg/L 0 1 2 3 4 5 Influent Effluent Figure 7. Changes in total inorganic nitrogen (TIN) and orthophosphate concentration in the influent and the effluent of the photobioreactors. Effluent values are the averages of the three photobioreactors. In mid-summer photobioreactors were stopped. Similar to TIN, the concentration of orthophosphate in the influent was lower than that considered during the design phase of the system (see Methods section in Supplementary saterials). In fact, it was not detected in a considerable number of sampling events and the calculated removal efficiency (when influent values were detected) was very high during all the study, greater than 95%, showing not clear seasonal trends for efficiency as observed for TIN. This high removal was achieved due to the uptake by microalgae, but also to their indirect effect on rising pH and causing ortophosphate precipitation (García et al., 2002). Reasons behind the lower influent TIN and orthophosphate concentrations than those considered for design are not evident, since data used during the design phase were based on field measurements. Average influent N:P ratio was 14.5:1 (in a milligram basis), slightly higher than that considered in design phase (7:1, see Methods section in the Supplementary Materials). This high ratio and the low phosphorus concentrations led mostly to a constant development of populations of Cyanobacteria in the PBRs, which were clearly dominant during summer and autumn. Most of the Cyanobacteria belonged to acoccal species resembling Synechococcus (Figure S4). Note that during these seasons, the influent N:P ratio was even higher than 20:1. In winter and a part of spring, however, Cyanobacteria were outcompeted by green microalgae. Microalgae of the inoculum were not observed in the PBRs after few months of operation. 3.2. Biomass production and final effluent quality Microalgae biomass concentration and production was measured through solids analyses. Note that microalgae evaluation through pigment measurements (chlorophyll a) probably would give a more accurate measurement of biomass concentrations. However, in this work solids were measured for the sake of simplicity and because in these microalgae systems there is a very good correlation between chlorophyll a and solids (García et al., 1998, 2006). Also because microalgae production in culture systems is usually given in biomass weight (Ansari et al., 2017). Solids concentrations were higher in the effluent (mixed liquor without biomass separation) than in the influent due to microalgae growth. In the effluent, average VSS represented the 78% of the TSS considering aggregated data of the three PBR. This percentage fits well into the values generally observed for microalgae based treatment systems (Gutiérrez et al., 2016), which are lower than other biomasses due to the high pH achieved in the mixed liquor, which promotes precipitation of inorganic salts of different nature. Biomass concentration showed great variations in the three PBRs, as evidenced by the high VSS standard deviation. For this reason, changes in the biomass concentration have been represented in box-plot charts for the different seasons (Figure 8). As observed, the biomass concentration is well related with environmental factors such as solar radiation and temperature. Considering the average concentration, the biomass production in the three PBR together was approximately 670 g VSS/d, which in volumetric terms represents approximately 20 g VSS/m3.d (=20 mg VSS/L·d). Obviously, production greatly changed between seasons with much lower values in winter (240 g VSS/d, 7 g VSS/m3·d) than in summer (1500 g VSS/d, 43 g/m3·d). García-Galán et al. (2018) obtained very similar biomass production rates in a similar PBR treating the same wastewater. These authors registered the highest production in April (approximately 75 g TSS/m3·d). Considering the VSS/TTS ratio found in the present work, the estimated production achieved by García-Galán et al. (2018) would be 57 g VSS/m3·d, which is in the same orders of magnitude to the values observed in this study. The great variation of biomass concentration (and production) observed in spring (Figure 9) was linked to the atypical cold and rainy weather in part of February, entire March and part of April 2018. Regarding the values observed for TIN and orthophosphate (Figure 6), it is quite clear that the biomass production was limited by nutrients during the studied period, except for winter and part of spring (when TIN concentration increased). Indeed, the environmental factors in corresponding months (solar radiation and temperature) are those that limit microalgae growth. Summer Autumn Winter Spring VSS, mg/L 0 100 200 300 400 500 600 Figure 8. Box-plot of the average volatile suspended solids (VSS) concentration in the three photobioreactors in different seasons. The lower boundary of the box indicates the 25th percentile, the line within the boxmark the median (solid line), and the upper boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentiles, respectively. Upper and bottom dots represent the 95th and 5th percentile, respectively. Influent COD values were relatively low due to the feedstock origin (mixture of agricultural run-off and treated domestic wastewater). PBR dissolved COD was slightly higher than total influent COD. This is due to the fact that, in these type of systems, microalgae biomass releases dissolved organic carbon (García et al., 2006). However, the impact of this increase in COD was irrelevant, as COD was clearly lower than the legal discharge requirements usually employed for secondary effluents (< 125 mg/L), together with the fact that TSS values in the final effluent were very low. The final effluent was produced after microalgae biomass separation in the static lamella settler, which main task was water clarification. One of the main aims of the INCOVER project was to produce clarified water with a very low turbidity (< 5 UNT), in order to prevent the good performance of the units after the PBRs (not described in this work, but shown in Figure 1 and S1). Figure 9 shows the changes in turbidity values and TSS concentration. As it can be observed, the quality of the final effluent was very good, values being most of the time below 5 UNT, and only with few peaks exceeding it. These peaks were related to small failures in plant operation, such as coagulant depletion or inappropriate coagulant doses that had to be re-establish after jar-test evaluation. Note that coagulant dosed usually ranged from 2 to 10 mg Al3+/L. The good quality in terms of turbidity was also reflected in the TSS concentrations, which were in all cases well below 25 mg/L, and in more than half of the cases below 15 mg/L. Note that dissolved constituents were scarcely changed from the PBR to the effluent of the settling tank. Time, months Dec Jan Feb Mar Apr May Turbidity, NTU TSS, mg/L 0 5 10 15 20 25 30 Turbidity TSS Figure 9. Changes in turbidity and TSS concentration after clarification (final effluent). Measurements from December 2018 when the settling tank was put in operation and optimized. 4. Conclusions The present study demonstrates the efficiency of microalgae based technologies such as PBRs to remove nutrients from wastewater at full scale. Data of one year of operation of 3 full-scale semi-closed horizontal tubular PBRs (11.7 m3) have been summarized and evaluated. These PBRs were conceived within the INCOVER project, with the dual purpose of removing nutrients from wastewater and producing biomass, which would be further valorized within a biorefinery concept and the circular economy framework. A daily flow of 2.3 m3/d per PBR was treated, corresponding to a HRT of 5 d. Microalgae activity has a strong dependency on seasonal changes in solar radiation and temperature, which led to higher pH values in the mixed liquor detected in summer and great variations in biomass production between seasons, going from 7 g VSS/m3·d in winter to 43 g VSS/m3·d in summer. Biomass separation efficiency of the settler was very high, with turbidity and TSS values < 5 UNT and around 15 mg/L, respectively. Average nitrogen removal efficiency ranged from 50% in winter to 90% in summer. Phosphorus concentration was much lower than that of nitrogen in both influent and the clarified effluent, mostly corresponding to orthophosphate, and the removal efficiency was greater than 95% in almost all cases. Acknowledgements The authors would like to thank Mr. Javier Carretero and the environmental analysis laboratory of the GEMMA-UPC group for their uninterested help and support during this study. Also the authors would like to thank the European Commission [INCOVER, GA 689242] for their financial support. M.J. García and E. Uggetti would like to thank the Spanish Ministry of Industry and Economy for their research grants [FJCI-2014-22767 and IJCI-2014-21594, respectively]. References Ansari, F.A., Singh, P., Guldhe, A., Bux, F., 2017. Microalgal cultivation using aquaculture wastewater: Integrated biomass generation and nutrient remediation. Algal Res. 21, 169–177. https://doi.org/10.1016/j.algal.2016.11.015 APHA-AWWA-WPCF, 1999. Standard methods for the examination of water and wastewater, 20th edition. Ávila, C., Garfí, M., García, J., 2013. Three-stage hybrid constructed wetland system for wastewater treatment and reuse in warm climate regions. Ecol. Eng. 61, 43–49. https://doi.org/10.1016/j.ecoleng.2013.09.048 Bohutskyi, P., Chow, S., Ketter, B., Fung Shek, C., Yacar, D., Tang, Y., Zivojnovich, M., Betenbaugh, M.J., Bouwer, E.J., 2016. Phytoremediation of agriculture runoff by filamentous algae poly-culture for biomethane production, and nutrient recovery for secondary cultivation of lipid generating microalgae. Bioresour. Technol. 222, 294–308. https://doi.org/10.1016/j.biortech.2016.10.013 García, J., Mujeriego, R. and Hernández-Mariné, M., 1998. Tratamiento de aguas residuales urbanas mediante lagunas de alta carga: evaluación experimental. Ingeniería del Agua 5, 35–50. https://doi.org/10.4995/ia.1998.2748 García-Galán, M.J., Gutiérrez, R., Uggetti, E., Matamoros, V., García, J., Ferrer, I., 2018. Use of full-scale hybrid horizontal tubular photobioreactors to process agricultural runoff. Biosyst. Eng. 166, 138–149. https://doi.org/10.1016/j.biosystemseng.2017.11.016 García, J., Green, B.F., Lundquist, T., Mujeriego, R., Hernández-Mariné, M., Oswald, W.J., 2006. Long term diurnal variations in contaminant removal in high rate ponds treating urban wastewater. Bioresour. Technol. 97, 1709–1715. https://doi.org/10.1016/j.biortech.2005.07.019 García, J., Hernández-Mariné, M., Mujeriego, R., 2002. Analysis of key variables controlling phosphorus removal in high rate oxidation ponds provided with clarifiers. Water SA 28, 55–62. Garcia, J., Mujeriego, R., Hernandez-Marine, M., 2000. High rate algal pond operating strategies for urban wastewater nitrogen removal. Appl. Phycol. 12, 331–339. https://doi.org/10.1023/a:1008146421368 García, J., Rousseau, D.P.L., Morató, J., Lesage, E., Matamoros, V., Bayona, J.M., 2010. Contaminant removal processes in subsurface-flow constructed wetlands: A review. Crit. Rev. Environ. Sci. Technol. https://doi.org/10.1080/10643380802471076 Gutiérrez, R., Passos, F., Ferrer, I., Uggetti, E., García, J., 2016. Harvesting microalgae from wastewater treatment systems with natural flocculants: effect on biomass settleability and biogas production. Algal Res. 9, 204-2011. https://doi.org/10.1016/j.algal.2015.03.010 Gutiérrez, R., Ferrer, I., González-Molina, A., Salvadó, H., García, J., Uggetti, E., 2016. Microalgae recycling improves biomass recovery from wastewater treatment high rate algal ponds. Water Res. 106, 539–549. https://doi.org/10.1016/j.watres.2016.10.039 Kamyab, H., Din, M.F.M., Keyvanfar, A., Majid, M.Z.A., Talaiekhozani, A., Shafaghat, A., Lee, C.T., Shiun, L.J., Ismail, H.H., 2015. Efficiency of Microalgae Chlamydomonas on the Removal of Pollutants from Palm Oil Mill Effluent (POME), in: Energy Procedia. pp. 2400–2408. https://doi.org/10.1016/j.egypro.2015.07.190 Kumar, A., Priyadarshinee, R., Roy, A., Dasgupta, D., Mandal, T., 2016. Current techniques in rice mill effluent treatment: Emerging opportunities for waste reuse and waste-to-energy conversion. Chemosphere. https://doi.org/10.1016/j.chemosphere.2016.08.118 Labbé, J.I., Ramos-Suárez, J.L., Hernández-Pérez, A., Baeza, A., Hansen, F., 2017. Microalgae growth in polluted effluents from the dairy industry for biomass production and phytoremediation. J. Environ. Chem. Eng. 5. https://doi.org/10.1016/j.jece.2016.12.040 Lananan, F., Abdul Hamid, S.H., Din, W.N.S., Ali, N., Khatoon, H., Jusoh, A., Endut, A., 2014. Symbiotic bioremediation of aquaculture wastewater in reducing ammonia and phosphorus utilizing Effective Microorganism (EM-1) and microalgae (Chlorella sp.). Int. Biodeterior. Biodegrad. 95, 127–134. https://doi.org/10.1016/j.ibiod.2014.06.013 Ledda, C., Tamiazzo, J., Borin, M., Adani, F., 2016. A simplified process of swine slurry treatment by primary filtration and Haematococcus pluvialis culture to produce low cost astaxanthin. Ecol. Eng. 90, 244–250. https://doi.org/10.1016/j.ecoleng.2016.01.033 Matamoros, V., Rodríguez, Y., 2016. Batch vs continuous-feeding operational mode for the removal of pesticides from agricultural run-off by microalgae systems: A laboratory scale study. J. Hazard. Mater. 309, 126–132. https://doi.org/10.1016/j.jhazmat.2016.01.080 Selman, M., Greenhalgh, S., Diaz, R., Sugg, Z., 2008. Eutrophication and hypoxia in coastal areas: a global assessment of the state of knowledge. WRI Policy Note 1–6. Solimeno, A., Gabriel, F., García, J., 2017. Mechanistic model for design, analysis, operation and control of microalgae cultures: Calibration and application to tubular photobioreactors. Algal Res. 21, 236–246. https://doi.org/10.1016/j.algal.2016.11.023 Solorzano, L., 1969. Determination of ammonia in natural waters by the phenolhypochlorite method. Limnol. Oceanogr. https://doi.org/10.4319/lo.1969.14.5.0799 Steffen, W., Richardson, K., Rockstrom, J., Cornell, S.E., Fetzer, I., Bennett, E.M., Biggs, R., Carpenter, S.R., de Vries, W., de Wit, C.A., Folke, C., Gerten, D., Heinke, J., Mace, G.M., Persson, L.M., Ramanathan, V., Reyers, B., Sorlin, S., 2015. Planetary boundaries: Guiding human development on a changing planet. Supplementary Material. Science (80-. ). 347, 1259855–1259855. https://doi.org/10.1126/science.1259855 Uggetti, E., García, J., Älvarez, J.A., García-Galán, M.J., 2018. Start-up of a microalgaebased treatment system within the biorefinery concept: from wastewater to bioproducts. Water Sci. Technol. In press. Uggetti, E., Hughes-Riley, T., Morris, R.H., Newton, M.I., Trabi, C.L., Hawes, P., Puigagut, J., García, J., 2016. Intermittent aeration to improve wastewater treatment efficiency in pilot-scale constructed wetland. Sci. Total Environ. 559, 212–217. https://doi.org/10.1016/j.scitotenv.2016.03.195