Partitioning of wastewater treatment high rate algal

Algal Research 19 (2016) 77–85

Contents lists available at ScienceDirect

Algal Research journal homepage: www.elsevier.com/locate/algal

Partitioning of wastewater treatment high rate algal pond biomass and particulate carbon Karl.A. Safi ⁎, Jason.B.K. Park, Rupert.J. Craggs National Institute of Water and Atmospheric Research, P.O. Box 11-115, Hamilton, New Zealand

a r t i c l e

i n f o

Article history: Received 20 March 2016 Received in revised form 17 July 2016 Accepted 27 July 2016 Available online xxxx Keywords: Algae Bacteria Biomass Cell carbon Detritus Microzooplankton

a b s t r a c t The partitioning of algae, bacteria, grazers and detritus in two wastewater treatment high rate algal ponds (HRAPs) was investigated in relation to particulate carbon (PC) over a year. Algae dominated the pond biomass accounting for ~61% of the PC. Changes in algal biomass correlated with changes in PC with both varying seasonally and having spring or summer maxima. The algal biomass itself was dominated by larger cells or colonies in the 20–200 μm size fraction with Pediastrum sp. prominent. Bacterial biomass, in contrast, only accounted for ~13.5% of the PC and varied less seasonally. Grazer biomass was lowest at ~4% of the PC on average and was dominated by either zooplankton or microzooplankton. Grazer biomass however, varied the most and reached ~14% of the PC during a spring zooplankton bloom that markedly reduced algal biomass. The remaining ~21.5% of the PC was made up of dead algal, detrital matter, and mucilage that tended to aggregated into bio-floccs. The PC of an efficiently operating HRAP is shown to be driven by high algal biomass with low bacterial and grazer biomass. If this balance is lost grazers may grow to levels that enable them to reduce algal biomass and productivity compounding pond instability. © 2016 Published by Elsevier B.V.

1. Introduction High rate algal ponds (HRAPs) are a promising technology for wastewater treatment as they provide improved treatment over conventional wastewater stabilisation ponds and have potential biofuel production applications [1–3]. HRAPs are shallow (~0.3 m deep) raceway ponds with gentle mixing (mean horizontal water velocity ~ 0.15–0.3 m/s) typically provided by a paddlewheel [4,5]. These ponds are specifically designed to promote algal growth and assimilation of soluble carbon and nutrients from wastewater. The oxygen produced during photosynthesis also promotes the breakdown of organic compounds by heterotrophic bacteria [5]. Therefore, HRAPs have similarities to well mixed super-eutrophic natural waterbodies as both favor microalgae production. Different environmental (light and temperature), operational (mixing, DO, pH, CO2 and nutrients) and biological (zooplankton grazers and algal pathogens) conditions influence the standing crop of HRAP biomass and overall productivity. The fixed or particulate carbon (PC) in HRAPs is contained within the various organic components including algae, bacteria, protozoans (ciliates and flagellates), zooplankton (rotifers and cladocerans), and other organic matter (e.g. dead algal cells, detrital matter, and mucilage). The component composition of the biomass, in turn, significantly influences productivity, settleability, and PC. Recent pilot-scale studies have focused on optimizing HRAP design and operation to achieve maximum algal production, algal biomass dominance ⁎ Corresponding author. E-mail address: karl.safi@niwa.cri.nz (K.A. Safi).

http://dx.doi.org/10.1016/j.algal.2016.07.017 2211-9264/© 2016 Published by Elsevier B.V.

and optimal harvest composition [3,4,6,7]. Under optimal conditions HRAP biomass will be dominated by typical eutrophic water algal species growing as large settleable colonies (colony diameter of 50–200 μm), that form large bio-floccs (diameter: N 500 μm) associated with bacteria and other organic matter [8]. The dominance of settleable colonial algae is desirable in HRAP's as it improves biomass harvest efficiency. In contrast ponds that are operating inefficiently may have low algal biomass; be dominated by small non-flocculating single celled algae; or have high bacterial or grazer biomass [4]. In natural waterbodies algae populations are controlled by grazers which are predated on by fish and other higher level predators, transferring algal biomass/carbon through the food web to higher trophic levels. HRAP algal biomass and production are maximized when grazer populations and grazing pressure are low. However, the lack of fish and other predators in HRAPs, means that they are susceptible to grazing by “blooms” of herbivorous zooplankton (e.g. rotifers and cladocerans) or protozoans (e.g. amoebae, ciliates and flagellates) particularly when small algae dominate. While some common zooplankton species (e.g. rotifers) have a preference to graze on small algae, ~8 μm [9], larger zooplankton (e.g. Moina sp.) can graze even the large colonial algae in HRAPs [10]. For example, a previous pilot-scale HRAP study [3] reported that algal biomass concentration was substantially reduced from ~250 to b 20 mg/L within 4 days when a population of Moina sp. increased above ~500 individuals/L. Large Pediastrum sp. colonies (~20–200 μm) dominated the algal biomass at this time indicating that grazing pressure by Moina sp. was effective even on large colonial algae. Since large zooplankton grazers have much slower growth rates than algae

78

K.A. Safi et al. / Algal Research 19 (2016) 77–85

[11] they can only affect the algal population if given sufficient time to increase their numbers. In contrast, protozoan grazers are capable of growing at similar rates to the smaller (b20 μm) algae on which they graze [12], which may be one reason why smaller algae are less prevalent in HRAPs. Protozoans also graze on bacteria which may assist in keeping bacteria from dominating pond biomass [13]. In addition to grazing HRAPs are also vulnerable to other biological influences such as fungal parasitism and viral infection which can also significantly reduce the algal population within a few days and trigger changes in algal population structure, diversity and succession [14]. A technique to further understand the relationships between organisms and their role in natural water body food-webs is to partition them in terms of their carbon biomass [15–17]. These studies have led to generalizations about the relative importance of different planktonic food web components such as: grazing versus microbial pathways; metazoan versus protozoan grazers; and algae versus bacteria as carbon sources in natural systems. The relative contribution of different organisms to the biomass of wastewater treatment HRAPs and other manmade super-eutrophic systems has not previously been quantified. There is limited information available on the partitioning of food-web components within HRAPs and the relative contribution of each component to PC has not been studied before. The main purpose of this study was to investigate over one year the changes in the relative contribution of the different biomass components to PC in wastewater treatment HRAPs. Specifically we aimed to; (i) identify the size classes of organisms and relative contribution to PC; (ii) identify the algal, bacterial, grazer and other components of HRAP biomass and determine their relative contribution to PC; (iii) describe any seasonal changes, and; (iv) identify the effects of grazing on the pond biological community. 2. Materials and methods 2.1. Experimental pilot-scale high rate algal ponds The experiment was conducted over one year (from April 2009 to March 2010) using two identical pilot-scale single-loop raceway HRAPs treating domestic wastewater at the Ruakura Research Centre, Hamilton, New Zealand (37o47′S, 175°19′E). Each HRAP had a surface area of 31.8 m2, a depth of 0.3 m and a total volume of 8 m3 with semi-circular end walls and with a dividing baffle separating the two raceway channels. The pond water was circulated around each raceway at a mean velocity of ~0.15 m/s by a paddlewheel. More detailed specifications of the HRAPs are described in a previous study [7]. The HRAPs were operated at different hydraulic retention times (HRT) depending on season to account for changes in environmental parameters such as light and temperature and their influence on wastewater treatment and algal growth. Algal production was maintained throughout the year by altering the HRAP HRT from 8 days in winter (June to August 2009) to 4 days in summer (November 2009 to March 2010) by using inflows of 1 and 2 m3/day respectively. In summer the influent sewage was diluted 1:1 with tap water to reduce nutrient concentration while maintaining the same nutrient load. During the NZ autumn (March–May 2009) and throughout spring (Sept–Nov 2009) the HRT of the HRAPs was maintained at 6 days by using an inflow of 1.3 m3/day of diluted sewage. CO2 was added to the HRAPs during the day to maintain the maximum pH below 8.0 to avoid carbon limitation and free ammonia inhibition [6]. Settled biomass was removed daily from an algal harvester (with an HRT of 3–6 h depending on season) and a portion of the gravity harvested algae (Pediastrum sp. dominant) was recycled back to one of the HRAP (HRAPr with recycling) at a recycling ratio of ~10% of daily algal production. The second HRAP was operated as a control without recycling (HRAPc) with all other operational parameters the same as the HRAPr. These operational differences were designed to maintain the dominant algal species, as well as the biomass levels, productivity and settleability of algal populations within

the HRAPs. Further details are described in previously published work using these HRAPs [3,4]. 2.2. Sampling, physical and chemical analysis including carbon Pond water physical properties such as dissolved oxygen (DO), pH and water temperature were recorded at 15 min intervals using a multiprobe DataSonde® (Hydrolab, HACH Environment, USA) coupled with a datalogger (CR10X, Campbell Scientific Inc., UT, USA) (Fig. 1). Daily sunlight radiation, air temperature, and sunshine hours were downloaded from the NIWA climate database for the Ruakura NIWA/ AgResearch Station, Hamilton, New Zealand (37o84′70″S, 175o81’90″ E), (http://cliflo-niwa.niwa.co.nz/), (Fig. 1). Samples of HRAP water (1 L) were taken at monthly intervals for one year from April 2009 to March 2010. 100 ml subsamples were taken from the 1 L sample and preserved with Lugol's Iodine solution (1% final concentration) for the analysis of algae, microzooplankton, zooplankton and an estimate of detritus (i.e. large bio-flocs containing dead algal cells, other detrital organic matter and mucilage). Two millilitre subsamples were also taken from the 1 L sample for the analysis of picophytoplankton and bacteria by either flow cytometry or for direct microscopic exanimation. Additional subsamples were taken for duplicate analysis of Chlorophyll-a (Chl-a) and particulates which were analysed following Standard Methods [18]. 2.3. Cell counting, biovolume calculation, and carbon conversion 2.3.1. Phytoplankton Up to 10 ml of the Lugol preserved subsamples was examined using a Leica inverted microscope (DMI 3000 B) to identify and count N2 μm size algae (Table 1). For enumeration, samples were settled in Utermöhl chambers for N4 h, and then identified and counted with the microscope at 100 × to 600 × magnification. The dimensions of each taxon were measured (except Pediastrum spp.) and the biovolume estimated from approximated geometric shapes (spheres, cones, ellipsoids) following [19,20]. The calculated biovolumes of algae were then used to determine algae cell carbon (μg C/L) using the conversion equations of [21] for different algal groups (Green algae, Diatoms, Dinoflagellates; see Table 1). Biovolumes of the colonial algae Pediastrum spp. were determined using a microscopic image analysis technique and details of this method are given in a previous publication [22]. Images were taken with a Leica microscopic camera (DFS 420c) and the dimensions (length and width) of each colony were measured using microscopic image analysis software (Leica Application Suite, LAS version 3.1.0) Pediastrum cell carbon was calculated by dividing the measured average colony biovolume by the numbers of cells per colony (8, 16, 32, 64). This was then converted to carbon using the regression equation for chlorophytes (green algae, see Table 1) [21]. Picophytoplankton were examined in duplicate either using Flow Cytometry following the methods [23], or direct counts made using autofluorescence [24]. Cell carbon for eukaryotic picophytoplankton was determined by estimating the average spherical diameter and converting to yield a factor of 820 fg C/cell, while for the cyanobacteria, Synechococcus type sp. a factor of 250 fg C/cell was used [12]. 2.3.2. Bacteria Duplicate bacterial samples (~2 ml) were frozen in liquid nitrogen and thawed immediately before counting. Initially counts by flow cytometry underestimated bacterial numbers due to clumping, attachment to mucilaginous cell surfaces and inclusion in aggregated matter even after sonification. To overcome these issues, samples were examined under the microscope after staining with Acridine Orange [29]. Bacteria cell carbon was then calculated using a conversion factor of 108 fg C/cell [30]. This was based on the measured average cell biovolume. As the bacteria are being cultured in a medium that is high in dissolved organic matter they have a larger average cell size than


79

Fig. 1. Changes in daily sunlight radiation (A) recorded near the pond site between April 2009 and April 2010 and Pond temperature (B), dissolved oxygen (C) and pH (D) for both high rate algal ponds (HRAP) between April 2009 and April 2010. (HRAPr = recycled, HRAPc = control).

typical marine and freshwater bacteria. In fresh and marine waters a large percentage of bacteria is often dormant and the cell size is therefore reduced.

2.3.3. Heterotroph grazer biomass The heterotroph grazer biomass consisted of ciliates and zooplankton including rotifers and cladocerans. An inverted microscope was used to enumerate and identify ciliates. Ciliate biovolumes were then determined using the same method reported for algae but a larger sample volume was measured at lower magnification to insure enough cells were counted (Table 1). The estimated ciliate biovolumes were then converted to biomass carbon using a factor of 0.19 pg C/μm3 [25]. Zooplankton were identified to genus or species level where possible and biomass carbon was calculated using the body length to carbon ratios for rotifers, Daphnia sp. and Moina sp. [27]. Copepods that were observed were given a carbon content of 2.8 μg C per individual [28].

2.4. Determination of organic and inorganic carbon Particulate carbon (PC) was determined by filtering subsamples onto pre-combusted GF/F filters followed by analysis using a carbon-hydrogen-nitrogen (CHN) analyzer (CE Instruments NC2500 with machine precision of ~ 2%). Blanks (GF/F filters) were also analysed and all filters were only handled by filter forceps to prevent any contamination. Once all measured organisms were converted to carbon (μg/L), the remainder was designated as unaccountable PC.

2.5. Calculated carbon groupings defined Biological carbon was calculated for individual groups as listed in Table 1. In addition, for comparison with directly measured PC, biological carbon was further grouped into Carbon Algal Biomass (CAlgalB) which was calculated by summing the converted carbon for each of the phytoplankton groups listed in Table 1 (N 2 μm + Pediastrum spp.

80


Table 1 Details and references for the methods used for identification, cell counting, biovolume calculation and carbon conversion of algae, bacteria and zooplankton in two wastewater high rate algal ponds (HRAP) during a one year experimental period. The cellular carbon content was determined from calculated cell volume and the listed carbon conversion factors using the equation; log pg C cell−1 = loga + b ∗ log V (μm3), where loga is the y-intercept and b is the slope for Phytoplankton N2 μm including Pediastrum spp. Other organisms were converted on a per cell basis using the listed carbon conversion factors. Organisms

Identification and cell counting methods

Phytoplankton N2 μm

Pediastrum spp. (20–200 μm) Picophytoplankton (b2 μm) Bacteria

Zooplankton

Ciliates

Calculation of cell biovolume

Sample volumes

Methods

2–10 ml Lugol preserved samples

Individual cell counting Based on geometric shapes using inverted microscope (e.g. sphere, rod, oval, cone, etc.)

1 ml of fresh pond water samples 2 ml frozen samples in liquid nitrogen 2 ml frozen samples in liquid nitrogen 5–20 ml

Microscope image analysis Flow Cytometry and/or autofluorescence

Average colony biovolume divided by the cell numbers Based on estimated ave. spherical diameter

Acridine Orange counting method

Based on estimated ave. spherical diameter

Individual counting using inverted microscope

Based on geometric shapes

Zooplankton (rotifers, Daphnia sp., Moina sp.) Copepods

20–200 μm + b2 μm picophytoplankton). “Total measured biological carbon” (MBC) was also calculated and is defined as the CAlgalB added to all other measured sources of carbon from biological organisms as listed in Table 1 (i.e. CAlgalB + bacteria + zooplankton).

Carbon conversion

Reference

Used equations of for different algal groups chlorophytes, (Loga − 1.026,b1.088) Diatoms (Loga − 0.541,b0.811) Dinoflagellates (Loga − 0.353,b0.864) Used regression equation for chlorophytes (Loga − 1.026,b1.088) 820 fg C/cell and 250 fg C/cell for Synechococcus type sp.

[19–21]

[21,22] [12,23]

Cell counts × 108 fg C/cell

[12,20]

Biovolume (μm3) × 0.19 pg C/μm3 ×

[25]

total counts Using body length to carbon ratios of each organisms Total counts × 2.8 μg C/count

[26,27] [28]

performance of the these HRAP's in terms of organic matter (TSS/VSS and BOD5) and nutrient (N and P) removal have been described previously [4,6,7]. 3.2. Relative contribution of biological organisms to particulate carbon (PC)

3. Results and discussion 3.1. Physical and chemical changes in pond water and influent water characteristics As expected solar radiation and pond temperature varied following the annual seasonal cycle (Fig. 1). dissolved oxygen was always high but peaked in the spring of 2009 and remained high through the summer of 2010 (Fig. 1). pH was maintained largely between 6 and 8 but did decline below 6 for a time in late July, early August of 2009 and was highest in January 2010 in HARPc (Fig. 1). While BOD and nutrient concentrations of the influent sewage were not continuously monitored during the experimental period, the influent sewage typically contained ~66 mg/L of BOD (47 mg/L of soluble BOD), 39 mg/L of ammoniacal-N and 5.4 mg/L of DRP on average. Further detailed wastewater treatment

Chlorophyll-a (Chl-a) has been traditionally used as the measure of algae biomass in HRAPs [3,4,30]. We investigated determining the relationship between PC and algal carbon using Chl-a converted into carbon content as has been reported in other carbon partitioning studies [15,17, 31,32]. Chl-a converted to carbon did show a strong positive correlation with PC and (p = 0.0001, r = 0.76, n = 20) but the relationship was consistently weaker than between directly calculated Algal Carbon (CAlgalB) and PC (p = 0.0001, r = 0.80, n = 20) so this measure was used to represent algal carbon for the remainder of this study. In HRAPr that had algal recycling, the proportion of PC that was CAlgalB varied between 39% and 77% over the experimental period (Fig. 2). MBC accounted for between 47% and 98% of PC over the experimental period (Fig. 2). Consequently MBC was more highly correlated with PC (p = 0.0001, r = 0.99, n = 11) than CAlgalB alone.

Fig. 2. Total measured particulate carbon (PC) compared to total measured biological carbon (MBC), (MBC was comprised of algae carbon + bacterial carbon + ciliate and zooplankton carbon) and compared to algae carbon (CAlgalB) alone. A = HRAPr, B = HRAPc. = PC; = MBC; □ = CAlgalB (±standard deviation).


81

Table 2 Average and seasonal percentage biomass of particulate carbon (PC) found in four biological carbon groupings in two high rate algal ponds (HRAP) over the period of one year between April 2009 and March 2010. HRAPr, HRAPc. (±standard deviation).

HRAPr Algae Grazer Bacteria Aggregated detritus HRAPc Algae Grazer Bacteria Aggregated detritus

Annual average

Winter

Autumn

Spring

Summer

62 ± 3.7 3 ± 11 13 ± 14 22 ± 10

48 ± 1.7 2 ± 1.8 20 ± 18 30 ± 15

69 ± 8.6 2 ± 2.4 17 ± 16 12 ± 22

64 ± 1.3 6 ± 6.6 11 ± 4.8 19 ± 9.8

68 ± 7.4 2 ± 1.7 5 ± 0.8 25 ± 8.8

56 ± 2 5 ± 3.7 12 ± 11 27 ± 18

38 ± 20 3 ± 3.0 19 ± 23 40 ± 15

69 ± 24 5 ± 3.4 8 ± 5.2 19 ± 22

65 ± 4.7 8 ± 6.8 10 ± 4.7 17 ± 15

61 ± 24 1 ± 0.8 10 ± 6.3 28 ± 20

The proportion of PC that was algal carbon (CAlgalB) in HRAPc (that had no algal recycling) varied more (ranging from 27% to 87%) than that in HRAPr (Fig. 2). The CAlgalB in HRAPc was also less strongly correlated with changes in PC (p = 0.003, r = 0.79, n = 11) but with the addition of other organisms the MBC accounted for between 43% and 99% of PC (in July and November) (Fig. 2). These results indicate that the variation of PC concentration in both HRAPs was strongly linked to changes in the relative abundance of pond biological organisms rather than other carbon sources (e.g. wastewater inputs). Algal biomass typically made up the majority of MBC and often PC, implying that the HRAPs were operating efficiently with high algal productivity. Algal carbon and consequently PC declined in autumn and winter, corresponding to the seasonal decline in HRAP algal productivity at lower light intensities and temperatures (Figs. 1, 2). During periods when algal biomass was low the contribution of other biological carbon sources to PC was consequently increased (Fig. 2). At times, almost all PC, could be accounted for by MBC, while at other times there was still an amount of unallocated PC. This was not unexpected given that the HRAPs are fed with waste organic matter, are designed to resuspend organic detrital matter, and form bio-flocs incorporating both living and dead organisms and other organic (and inorganic) substrates. The results indicate that even HRAPs that are performing well contain both “biologically active living organisms” and “detrital matter” in their PC. The unallocated PC represented by detrital matter was observed to be largely associated with the unmeasured components of the algalbacterial bio-floccs, which were commonly found in the HRAPs. The unmeasured components include unidentified organic detritus, dead algal cellular matter, zooplankton faecal pellets and extracellular mucilage compounds. Moreover, the extracellular mucilages were a habitat for

some of the measured bacterial populations as it provided both a source of carbon and a haven from direct grazing by protozoans. Given that these unmeasured sources of PC were mainly observed within aggregates or bio-flocs they were defined as “aggregated detritus” (AD). Such AD matter is commonly reported both in marine (marine snow) and freshwater environment's [33,34] but are more concentrated in the HRAPs due to both the high level of organic matter in the influent wastewater and the maintenance of floccs in suspension by the paddlewheel mixing. Large dead empty Pediastrum cells were also a common feature in bio-floccs and these cell walls are known to be resistant to decomposition and are even used as a tracers for historical algal populations in sediments [35]. During the one year experimental period of this study on average ~ 31% of PC was accounted for as AD in HRAPc, which was higher than that in the HRAPr (~ 22% on annual average) (Table 2). The variation of AD correlated with total PC in both HRAPr (p = 0.0001, r = 0.86, n = 11) and HRAPc (p = 0.003 r = 0.78, n = 11) but only weakly correlated with other measured biological carbon sources (CAlgaeB, Chl-a, MBC, grazers, bacteria) indicating that no one population consistently determined the proportion of AD in PC. 3.3. Seasonal variability in algal carbon The proportion of CAlgalB in the PC also varied seasonally. For example, CAlgalB made up the highest proportion of the PC in HRAPr during autumn and summer at 68 and 69% respectively but was lowest in winter at only 48%. A similar but more variable pattern occurred in HRAPc with the proportion of CAlgalB in the PC ranging from 69% in autumn to only 38% in winter (Table 2). These changes occurred despite the greater residence time of 8 days in winter, compared to only 4 days in

Fig. 3. Total Bacterial and Grazer carbon (Ciliate + Zooplankton) as measured in two high rate algal ponds (HRAP's) between April 2009 and March 2010. A = HRAPr, B = HRAPc (±standard deviation).

= bacterial C

= Grazer C,

82


Fig. 4. Zooplankton and ciliate carbon measured in two High rate ponds over between April 2009 and March 2010.

summer, suggesting again that seasonal limitations on algal productivity are the primary factor in overall pond PC. 3.4. Heterotroph carbon Of the heterotrophic organism's, bacterial carbon was the most important contributor to PC in the HRAPs, representing up to 45% of PC in HRAPc and 39% in HRAPr in June 2009 (Fig. 3). The bacterial carbon contribution had an almost opposite pattern to that of the algae carbon contribution, and was lowest in summer and autumn (HRAPr: 5%; HRAPc: 7.5%) and highest in winter (HRAPr: 30%; HRAPc: 19%) (Table 2). This suggests that when the HRAPs achieved high algal production in summer and autumn, bacterial numbers remained relatively stable (and low). This trend is the opposite of what would be seen in less eutrophic environments where bacterial populations mainly rely on algal exudate as a carbon source and increase as algal populations and production increase [36]. The more stable bacterial populations found in the current study probably reflect bacterial levels sustained by a high level of readily available dissolved organic carbon in the wastewater, but there were some exceptions where the bacterial population varied. Elevated bacterial numbers and thus high bacterial carbon found in HRAPc during June 2009 were due to high filamentous bacteria numbers. The cause of this increase is not clear but may reflect changes in wastewater organic matter or be triggered by fungal or viral effects on the pond biota [14]. Bacterial carbon in both HRAPs also increased (albeit slightly) around October/November 2009 after or during the period of zooplankton grazing. This may be related to the effects of an increased grazer population that may have resulted in an increase in the level of dissolved organic matter available for bacterial growth, both through sloppy feeding and reduced algal uptake [37]. Bacterial numbers and biomass in HRAPc also increased in January 2010 with no apparent direct biological cause. This could however, just reflect a change in wastewater organic matter or a lack of grazing as low ciliate biomass was reported at this time (Fig. 4). 3.5. Algal composition Algal biomass in HRAPr was dominated by Pediastrum sp. as a result of algal recycling [3,4] although this was largely varieties of Pediastrum boryanum other Pediastrum spp. including Pediastrum duplex were also present. The average contribution of Pediastrum to algal carbon during the summer was 94% and 70% in HRAPr and HRAPc respectively, but was only 32% of algal carbon in HRAPc during March 2010 when other smaller algae were dominant (Table 5). The other main algal species found in the HRAPs are summarized in Table 3. During the summer of

= Zooplankton,

= Ciliates, A = HRAPr, B = HRAPc.

2009 a mix of Desmodesmus, Scenedesmus, and Acutodesmus sp. was found in both HRAPs but only at very low (near detection limit) levels (Table 5). Micractinium sp. was also found in winter and spring in both HRAPs and was a major contributor to algal biomass in HRAPc dominating during October 2009 and contributing up to 38% of the algae carbon measured. A bloom of a diatoms (cf. Cyclotella sp.) occurred in HRAPc during March 2010, such that this small diatom (b20 μm) dominated the algal biomass contributing to 51% of the algae carbon measured (Table 5). A mix of unicellular coccoid algae and small flagellates (b20 μm) including Chlorella sp. and Chlamydomonas sp. were also consistently found in both HRAP's but they were larger contributors to algal carbon in the HRAPc. Coelastrum was the only other species to represent N 5% of the algal carbon during any month of the one year experimental period. As previously reported in [3,4] the results in the current study clearly show that HRAPc (without algal recycling) had a higher algal diversity and lower Pediastrum sp. dominance compared with that in HRAPr with algal recycling. A number of other species were found but these were always b5% of the overall carbon biomass. This group of other genera included picophytoplankton (b2 μm), which are a major contributor in some marine and freshwater environments [38]. They did not however, significantly contribute to algal carbon in the HRAPs in the current study, as their growth is not favoured in highly eutrophic well mixed environments [39]. Another cause of the low concentration of picophytoplankton may have been their higher susceptibly to grazers than the larger colonial algae (e.g. Pediastrum sp.), [39]. 3.6. Grazer composition and grazing Grazer composition (Table 4, Fig. 4) varied over the one year experimental period with both smaller ciliates and larger zooplankton important at times. The largest ciliate species, Paramecium sp. peaked during July and October 2009 in both HRAPc and HRAPr while a mix of smaller oligotrichs including Halteria sp. dominated ciliates for the rest of the year. Vorticella sp. and Epistylis sp. were prevalent during June and November 2009 as well as March 2010, while ciliates dominated grazer populations in both HRAPs from April through to August 2009. During spring (September to October 2009) larger zooplankton dominated in the HRAPs and are likely to have reduced the algal biomass at this time (Fig. 4). Rotifer (mainly Filinia logiseta and Brachionus spp.) and Cladoceran (Daphnia and Monia sp.) populations were elevated in spring and also present in November and December 2009 while copepods were only occasionally found (peaking during March 2010 in HRAPc). During January and February 2010 the pattern changed between the HRAPs with ciliates dominating in the HRAPr while

K.A. Safi et al. / Algal Research 19 (2016) 77–85 Table 3 Common algae species reported in two high rate algal ponds between April 2009 and March 2010. Trebouxiophyceae Actinastrum hantschii (Lagerheim 1882) Dictyosphaerium sp. Dictyosphaerium ehrenbergianum Nägeli 1849 Micractinium pusillum Fresenius 1858 Chlorella spp. Lagerheimia spp. Lagerheimia chodatii C.·Bernard 1908 Lagerheimia citriformis (Snow) Collins 1909 Chlorophyceae Ankistrodesmus spp. Ankistrodesmus falcatus (Corda ex Ralfs 1848) Ankistrodesmus fusiformis (Corda ex Korshikov) Chlamydomonas sp. Chlorogonium sp. Pandorina sp. Pandorina morum (O.F. Müller) Bory de Saint-Vincent 1824 Eudorina elegans Ehrenberg 1832 Coelastrum sp. Coelastrum cambricum var. cristata Kammerer Coelastrum microporum Nägeli in A. Braun1855 Desmodesmus spp. Desmodesmus intermedius (Chodat) E. Hegewald 2000 Acutodesmus sp. Acutodesmus acuminatus (Lagerheim) Tsarenko in Tsarenko & Petlovanny 2001 Pediastrum sp. Pediastrum boryanum var. longicorne f. glabra Raciborski Pediastrum boryanum var. cornutum Raciborski Pediastrum duplex Meyen 1829 Ankyra Ankyra judayi (G.M. Smith) Fott 1957 Conjugatophyceae Closterium sp. Closterium aciculare T. West 1860 Closterium acutum Brébisson in Ralfs 1848 Closterium moniliferum Ehrenberg ex Ralfs1848 Cosmarium sp. Cosmarium bioculatum Brébisson ex Ralfs1848 Euglenophyceae Euglena sp. Euglena acus Ehrenberg, 1830 Euglena gracilis Klebs, 1883 Trachelomonas volvocina (Ehrenberg) Ehrenberg 1834 Phacus sp. Phacus torta (Lemmermann) Skvortzov 1928 Dinophyceae Gymnodinium spp. Cryptophyceae Cryptomonas sp. Coscinodiscophyceae Cyclotella spp. Fragilariophyceae Synedra sp. Synedra acuta Ehrenberg 1843 Bacillariophyceae Achnanthes sp. Navicula sp. Gomphonema sp. Rhoicosphenia sp. Nitzschia sp. Synurophyceae Mallomonas spp.

zooplankton continued to dominate in the HRAPc. Peaks in ciliate populations and their domination of gazers in winter coincided with higher numbers of small sized (free living) algae and bacteria suggesting that these were their primary prey.

3.7. Contribution of different size classes of algae to particulate carbon (PC) The contribution of different size classes of algae to PC was investigated in both HRAPs (Table 5). Over the one year experimental period,

83 Table 4 Common zooplankton and ciliates found in two high rate algal ponds over one year from April 2009 and March 2010. Zooplankton Rotifera Brachionidae Brachionus spp. Brachionus rubens Pallas, 1766 Brachionus calyciflorus Pallas, 1766 Lepadellidae Lepadella sp. Trochosphaeridae Filinia longiseta Ehrenberg, 1834 Arhropoda Daphniidae Daphnia sp. Moinidae Moina sp. Cyclopidae Cyclopoid copepod Diaptomidae Calanoid copepod Others Copodida Nauplii Unidentified Nauplii Ciliates Parameciidae Paramecium spp. Vorticellidae Vorticella sp. Epistylididae Epistylis sp. Halteriidae Halteria sp. Oxytrichidae Histriculus sp. Others Unidentified Ciliophora Unidentified Oligotrichs

the majority of the algal biomass was in the 20–200 μm size fraction, with 94% (annual average) found in this size class in HRAPr and 78% in HRAPc. The 20–200 μm size class made up the largest proportion of algal biomass during summer in both HRAPr, and HRAPc, but made up the lowest proportion during winter in HRAPr, and autumn in HRAPc (Table 5). The 2–20 μm size class was the next most important fraction in terms of PC, and ranged between b1 and 13% of the carbon in HRAPr and between 4 and 61% in HRAPc depending on the dominant algal species in the ponds. The b2 μm (Picophytoplankton) size fraction was always low in the HRAPs at b 1 to 2.5% of the algal carbon. Algae in the N200 μm size fraction was also always low at b 1%. However, these values did not consider algae integrated into large bio-flocs (N 200 μm) in the HRAPs. Within the large bio-flocs we estimated that on average ~ 65% of their total biomass by area was algae although this was highly variable. Table 5 Average and seasonal percentage size class contributions of measured organisms to total PC in two high rate algal ponds over a period of one year between April 2009 and March 2010. HRAPr, HRAPc. (± standard deviation).

HRAPr b2 μm 2–20 μm 20–200 μm N200 μm HRAPc b2 μm 2–20 μm 20–200 μm N200 μm

Annual average

Winter

Autumn

Spring

Summer

14±11 3±7.7 81±14 2±5.8

21±5.2 5±18 74±18 b1±4.7

18±23 2±3.1 78±21 2±2.6

12±4.9 4±7.4 81±13.7 3±10.9

5±6.3 1±1.2 93±5.5 1±0.0

15±12 13±15 70±17 2±2.4

26±22.8 9±3.0 65±20 1±0.2

11±5.2 28±31 58±32 3±1.0

11±5.0 15±9.6 70±14 5±3.6

10±6.3 6±1.2 83±5.0 1±0.8

84


The size class distribution of all biological organisms including algae, bacteria and gazers was further investigated (Table 5). The algae dominated 20–200 μm fraction remained the dominant contributor to PC, but during winter the bacterial population considerably increased its contribution to the b 2 μm fraction from b2.5% of PC up to 21% and 26% in HRAPr and HRAPc respectively. Only zooplankton grazers were found in the N 200 μm size fraction, which was highest (8% of PC) during October 2009 in both HRAPr and HRAPc but was low throughout the rest of the year. Furthermore, the 2–20 μm fraction was low in both HRAPs (annual average: HRAPr: 3%; HRAPc: 13% of PC). In terms of functional size a large proportion of the bacteria was associated with algae or part of bio-floc aggregations found in the HRAPs. Thus, their proportional contribution to the large N200 μm aggregate fraction will have been underestimated in this study. In contrast, unicellular and flagellated algae were rarely a part of aggregates, suggesting that estimates of this size faction were more accurate. Since the majority of algal biomass and carbon was in the 20–200 μm size class, only larger zooplankton grazers would have been capable of grazing these colonies. Some grazers have specialised apparatus (parasitic, peduncular or engulfing) enabling them to graze on prey larger than themselves [40], but these were not commonly observed in the HRAPs. Moreover, as the large bio-floc aggregations incorporated a large proportion of bacteria and some smaller algal cells, a high proportion of the bacteria and smaller algae would not have been available to smaller free living grazers. This type of selection for larger algae reflects patterns observed in other eutrophic systems where formation of large colonies and aggregation are favoured strategies to reduce grazing pressure [41]. 4. Summary and conclusions Algal carbon in two pilot-scale HRAPs studied over a one year monitoring period varied seasonally and was highly correlated to PC suggesting that PC was largely determined by algae production and consequent carbon biomass. In contrast, the active heterotrophic components including bacteria and grazers never dominated PC. Bacteria was more significant in winter (up to 45% of PC in June) but this proportional increase only occurred because algal biomass was low and did not represent any large increase in bacterial biomass. A significant annual average ~ 21.5% of PC was attributed to aggregated detritus (AD) which was associated with large bio-floc aggregations and contained a mix of dead organic components derived from both external waste water inputs and dead algae and other dead organisms. This fraction of PC was often highest when biomass was elevated suggesting that at these times dead algae cells, other algal derived detritus and faecal pellets were important. Grazer populations and thus their carbon were the most highly variable component of PC, usually having low biomass (and PC) although at times, populations increased and there was clear evidence that grazing affected algal populations. Ciliates (Paramecium sp., Vorticella sp. and Epistylis sp.) were clearly identified as the dominant gazer biomass when small sized species of algae and bacteria (free living and a b 20 μm size class) were prevalent. This result implies that the small algal cells and bacteria were easily grazed by these Protista. The grazing of these small algal cells is also likely to have promoted the dominance of the larger sized colonial algae (e.g. Pediastrum sp.) observed in these HRAPs. However, when grazers were at their maximum abundance during a spring bloom they were dominated by larger zooplankton (including rotifers and cladocerans). At this time these larger grazers significantly reduced the population of even the larger algal colonies (2–200 μm size class including Pediastrum species), resulting in a decrease in algal carbon and consequently PC. The effects of grazers in general on PC and biomass was also partially mitigated by the integration of smaller algae and bacteria into larger bio-flocs (N200 μm), allowing grazing avoidance from, fast growing, smaller Protozoa. Overall this study shows that partitioning of PC into its biological

components allows us to begin to understand the biomass balance required to maintain a productive and efficient HRAP. Ideally a balanced HRAP will contain high algal biomass (N60% of PC) dominated by large colonial algae (N20 μm) with low bacterial biomass (b 20% of PC) and a low grazer biomass (b 3% of PC) dominated by protozoa (to maintain size class structure). It has also allowed us to identify some of the drivers of HRAP imbalance including excess grazing and algal size class/species changes, which can significantly influence the biomass and productivity of these systems.

Acknowledgements We would like to acknowledge Donna Sutherland and the unidentified reviewers for the helpful comments and review of this article and the NIWA Analytical Laboratory in Hamilton for analysis of chemistry samples. This research was funded by the New Zealand Ministry of Business, Innovation and Employment (Contract C01X0810).

References [1] Y. Chisti, Biodiesel from microalgae, Biotechnol. Adv. 25 (2007) 294–306. [2] R.J. Craggs, S. Heubeck, T.J. Lundquist, J.R. Benemann, Algal biofuels from wastewater treatment high rate algal ponds, Water Sci. Technol. 63 (2011) 660–665. [3] J. Park, R. Craggs, A. Shilton, Enhancing biomass energy yield from pilot-scale high rate algal ponds with recycling, Water Res. 47 (2013) 4422–4432. [4] J. Park, R. Craggs, A. Shilton, Recycling algae to improve species control and harvest efficiency from a high rate algal pond, Water Res. 45 (2011) 6637–6649. [5] A. Shilton, Pond Treatment Technology, IWA publishing, 2005. [6] J. Park, R. Craggs, Wastewater treatment and algal production in high rate algal ponds with carbon dioxide addition, Water Sci. Technol. 61 (2010) 633–639. [7] J. Park, R. Craggs, A. Shilton, Wastewater treatment high rate algal ponds for biofuel production, Bioresour. Technol. 102 (2011) 35–42. [8] J. Lee, D.-H. Cho, R. Ramanan, B.-H. Kim, H.-M. Oh, H.-S. Kim, Microalgae-associated bacteria play a key role in the flocculation of Chlorella vulgaris, Bioresour. Technol. 131 (2013) 195–201. [9] K. Rothhaupt, Differences in particle size-dependent feeding efficiencies of closely related rotifer species, Limnol. Oceanogr. 35 (1990) 16–23. [10] J. Benemann, P. Pedroni, Biofixation of fossil CO2 by microalgae for greenhouse gas abatement, Treccani Encyl. Hydrocarbons 3 (2007) 837–861. [11] D.J. Gifford, L.M. Fessenden, P.R. Garrahan, E. Martin, Grazing by microzooplankton and mesozooplankton in the high-latitude North Atlantic Ocean: spring versus summer dynamics, J. Geophys. Res. (C. Oceans 1978–2012) 100 (1995) 6665–6675. [12] K.A. Safi, F.B. Griffiths, J.A. Hall, Microzooplankton composition, biomass and grazing rates along the WOCE SR3 line between Tasmania and Antarctica, Deep Sea Res. (I Oceanogr. Res. Pap.) 54 (2007) 1025–1041. [13] W. Foissner, Taxonomische und okologische Revision der Ciliaten des Saprobiensystems, 1991. [14] M. Kagami, A. de Bruin, B.W. Ibelings, E. Van Donk, Parasitic chytrids: their effects on phytoplankton communities and food-web dynamics, Hydrobiologia 578 (2007) 113–129. [15] U. Gaedke, N. Kamjunke, Structural and functional properties of low-and high-diversity planktonic food webs, J. Plankton Res. 28 (2006) 707–718. [16] U. Gaedke, D. Straile, Seasonal Changes of the Quantitative Importance of Protozoans in a Large Lake: An Ecosystem Approach Using Mass-Balanced Carbon Flow Diagrams1994. [17] B. Auer, U. Elzer, H. Arndt, Comparison of pelagic food webs in lakes along a trophic gradient and with seasonal aspects: influence of resource and predation, J. Plankton Res. 26 (2004) 697–709. [18] A.P.H. Association, Water Environment Federation (1998) Standard Methods for the Examination of Water and Wastewater, Washington, DC, 1994. [19] E. Rott, Some results from phytoplankton counting intercalibrations, Schweiz. Z. Hydrol. 43 (1981) 34–62. [20] H. Hillebrand, C.D. Dürselen, D. Kirschtel, U. Pollingher, T. Zohary, Biovolume calculation for pelagic and benthic microalgae, J. Phycol. 35 (1999) 403–424. [21] S. Menden-Deuer, E.J. Lessard, Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton, Limnol. Oceanogr. 45 (2000) 569–579. [22] J.B. Park, R.J. Craggs, A.N. Shilton, Investigating the life-cycle and growth rate of Pediastrum boryanum and the implications for wastewater treatment high rate algal ponds, Water Res. 60 (2014) 130–140. [23] J. Hall, M. James, J. Bradford-Grieve, Structure and dynamics of the pelagic microbial food web of the subtropical convergence region east of New Zealand, Aquat. Microb. Ecol. 20 (1999) 95–105. [24] K.A. Safi, J.A. Hall, Factors influencing autotrophic and heterotrophic nanoflagellate abundance in five water masses surrounding New Zealand, N. Z. J. Mar. Freshw. Res. 31 (1997) 51–60. [25] M. Putt, D.K. Stoecker, An experimentally determined carbon: volume ratio for marine “oligotrichous” ciliates from estuarine and coastal waters, Limnol. Oceanogr. 34 (1989) 1097–1103.

K.A. Safi et al. / Algal Research 19 (2016) 77–85 [26] H.-M. Cauchie, L. Hoffmann, J.-P. Thomé, Metazooplankton dynamics and secondary production of Daphnia magna (Crustacea) in an aerated waste stabilization pond, J. Plankton Res. 22 (2000) 2263–2287. [27] I.V. Telesh, M. Rahkola, M. Viljanen, Carbon content of some freshwater rotifers, Hydrobiologia 387 (1998) 355–360. [28] J. Walve, U. Larsson, Carbon, nitrogen and phosphorus stoichiometry of crustacean zooplankton in the Baltic Sea: implications for nutrient recycling, J. Plankton Res. 21 (1999) 2309–2321. [29] S.F. Nishino, Direct acridine orange counting of bacteria preserved with acidified lugol iodine, Appl. Environ. Microbiol. 52 (1986) 602–604. [30] R.Y. Surampalli, R. Tyagi, Advances in Water and Wastewater Treatment, American Society of Civil Engineers, 2004. [31] H.J. Carrick, G.L. Fahnenstiel, E.F. Stoermer, R.G. Wetzel, The importance of zooplankton-protozoan trophic couplings in Lake Michigan, Limnol. Oceanogr. 36 (1991) 1335–1345. [32] G.L. Fahnenstiel, A.E. Krause, M.J. McCormick, H.J. Carrick, C.L. Schelske, The structure of the planktonic food-web in the St. Lawrence Great Lakes, J. Great Lakes Res. 24 (1998) 531–554. [33] G.A. Jackson, A.B. Burd, Aggregation in the marine environment, Environ. Sci. Technol. 32 (1998) 2805–2814.

85

[34] H.-P. Grossart, M. Simon, Bacterial colonization and microbial decomposition of limnetic organic aggregates (lake snow), Aquat. Microb. Ecol. 15 (1998) 127–140. [35] D.M. John, B.A. Whitton, A.J. Brook, The Freshwater Algal Flora of the British Isles: An Identification Guide to Freshwater and Terrestrial Algae, Cambridge University Press, 2002. [36] G. Bratbak, T. Thingstad, Phytoplankton-bacteria interactions: an apparent paradox? Analysis of a model system with both competition and commensalism, Mar. Ecol. Prog. Ser. 25 (1985) 23–30. [37] B.H. Hygum, J.W. Petersen, M. Søndergaard, Dissolved organic carbon released by zooplankton grazing activity - a high-quality substrate pool for bacteria, J. Plankton Res. 19 (1997) 97–111. [38] C. Callieri, J.G. Stockner, Freshwater autotrophic picoplankton: a review, J. Limnol. 61 (2002) 1–14. [39] C. Callieri, Picophytoplankton in freshwater ecosystems: the importance of smallsized phototrophs, Freshw. Rev. 1 (2008) 1–28. [40] S.L. Strom, Growth and grazing rates of the herbivorous dinoflagellate Gymnodinium sp. from the open subarctic Pacific Ocean, Mar. Ecol. Prog. Ser. 78 (1991) 103–113. [41] M. Lürling, E.V. Elert, Colony formation in Scenedesmus: no contribution of urea in induction by a lipophilic Daphnia exudate, Limnol. Oceanogr. 46 (2001) 1809–1813.