Académique Documents
Professionnel Documents
Culture Documents
Inuence of carbohydrate addition on nitrogen transformations and greenhouse gas emissions of intensive aquaculture system
Zhen Hu a, Jae Woo Lee b, Kartik Chandran c, Sungpyo Kim b, Keshab Sharma d, Samir Kumar Khanal e,
a
Shandong Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, Jinan 250100, China Department of Environmental Engineering, College of Science and Technology, Korea University, Sejong-ro 2511, Sejong 339-700, South Korea Department of Earth and Environmental Engineering, Columbia University, 500 West 120th Street, New York, NY 10027, USA d Advanced Water Management Centre, University of Queensland, St. Lucia, QLD 4072, Australia e Department of Molecular Biosciences and Bioengineering, University of Hawaii at Manoa, Honolulu, HI 96822, USA
b c
H I G H L I G H T S Addition of soluble starch to intensive aquaculture system enhanced heterotrophic bacterial growth and denitrication. Soluble starch addition minimized the nitrous oxide emissions from intensive aquaculture system by 83.4%. Soluble starch addition had signicant adverse effects in controlling GHG emissions from aquaculture systems.
a r t i c l e
i n f o
a b s t r a c t
Aquaculture is one of the fastest-growing segments of the food economy in modern times. It is also being considered as an important source of greenhouse gas (GHG) emissions. To date, limited studies have been conducted on GHG emissions from aquaculture system. In this study, daily addition of sh feed and soluble starch at a carbonto-nitrogen (C/N) ratio of 16:1 (w/w) was used to examine the effects of carbohydrate addition on nitrogen transformations and GHG emissions in a zero-water exchange intensive aquaculture system. The addition of soluble starch stimulated heterotrophic bacterial growth and denitrication, which led to lower total ammonia nitrogen, nitrite and nitrate concentrations in aqueous phase. About 76.2% of the nitrogen output was emitted in the form of gaseous nitrogen (i.e., N2 and N2O) in the treatment tank (i.e., aquaculture tank with soluble starch addition), while gaseous nitrogen accounted for 33.3% of the nitrogen output in the control tank (i.e., aquaculture tank without soluble starch addition). Although soluble starch addition reduced daily N2O emissions by 83.4%, it resulted in an increase of daily carbon dioxide (CO2) emissions by 91.1%. Overall, starch addition did not contribute to controlling the GHG emissions from the aquaculture system. 2013 Elsevier B.V. All rights reserved.
Article history: Received 12 July 2013 Received in revised form 12 September 2013 Accepted 17 September 2013 Available online xxxx Editor: Prof. P. Kassomenos Keywords: Intensive aquaculture system Nitrogen transformations Greenhouse gases Nitrous oxide Carbohydrate addition
1. Introduction Global food sh supply has increased dramatically in the last ve decades, with an average growth rate of 3.2% per year in the period of 19612009, to meet the rising world demands for protein. In 2010, capture sheries and aquaculture supplied about 148 million metric tons of sh globally (FAO, 2012). Because world shery productions have leveled off since the 1970s, aquaculture plays a crucial role in meeting the increasing global demands for sh. Since the expansion of aquaculture is restricted by land and water requirements, intensive aquaculture, which aims at raising sh at mass densities as high as 30 kg/m3 in closed or semi-closed systems with sufcient oxygen, fresh water and feed, has been increasingly
Corresponding author. Tel.: +1 808 956 3812; fax: +1 808 956 3542. E-mail address: khanal@hawaii.edu (S.K. Khanal). 0048-9697/$ see front matter 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.09.050
popular to overcome space and resource limitations (Delong et al., 2009). However, the development of intensive aquaculture has created serious environmental problems. The most important issue during the management of intensive aquaculture is avoiding the accumulation of toxic inorganic nitrogen species (especially, NH3 and NO 2 ) in the aqueous phase (Durborow et al., 1997; Hargreaves and Tucker, 2004). In intensive aquaculture systems, sh are fed a high protein diet with protein levels varying from 25 to 55% (Pillay and Kutty, 2005). Protein is digested by sh, producing mainly ammonia, and is excreted to the surrounding aqueous phase. In the aqueous phase, ammonia exists in two forms: un-ionized ammonia (NH3) and + ionized ammonium (NH+ 4 ). The sum of NH4 and NH3 is usually referred to as total ammonia nitrogen (TAN). Ammonia can further be oxidized to NO 2 and NO3 by nitrifying bacteria present in the aquaculture system. Inorganic nitrogen species, especially NH3 and NO 2 , are toxic to sh. High concentrations of NH3 and NO 2 can stimulate the release of corticosteroid hormones into the venous circulation, which may inhibit
194
sh growth and cause a variety of physiological dysfunctions (Tomasso, 1994). One of the most common approaches to prevent excess nitrogen buildup is through water exchange. This approach, however, has several environmental issues, e.g., requires a perpetual supply of fresh water and the generation of nitrogen-rich efuent, among others (Avnimelech, 2009). Another approach is to enhance nitrication to facilitate the conversion of ammonia and NO 2 into relatively non-toxic NO 3 by using nitrifying biolters. Although nitrifying biolters have been successfully employed in aquaculture systems and are effective in ammonia and NO 2 removal, they are costly and most importantly, do not provide an opportunity for nutrients recycling (Avnimelech, 2006; Crab et al., 2007). This is particularly signicant because protein in sh feed is the most expensive component of aquaculture operational costs especially when taking into consideration the fact that only about 25% of the nitrogen consumed by sh is converted to sh mass (Hargreaves, 1998). An additional strategy, namely carbohydrate addition, has been demonstrated to be effective in achieving low or zero-water exchange intensive aquaculture systems (Gao et al., 2012). By adding carbohydrate to aquaculture systems and regulating the carbon-tonitrogen (C/N) ratios, the growth of heterotrophic bacteria can be stimulated, which results in the removal of inorganic nitrogen through assimilation. As bacterial biomass increase, they tend to form noticeable aggregates (i.e., bioocs), which serve as a potential source of food for sh. Thus, the utilization of protein in the sh feed is enhanced due to nutrients recycling by heterotrophic bacteria. Recently, this approach has become increasingly popular for closedwater shrimp and tilapia cultivation (Asaduzzaman et al., 2010; Widanarn et al., 2012). Besides water pollution, the contribution of aquaculture systems to global greenhouse gases (GHG) emissions has aroused great attention in recent years (Williams and Crutzen, 2010; Hu et al., 2012). In aquaculture systems, major GHG emissions include carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O). CO2 is produced from the decomposition of organic materials and through respiration by sh. High levels of CO2 can be detected in aquaculture systems with high feed loading rates and relatively slow water turnover (Good et al., 2010). CH4 is produced when organic materials are broken down under anaerobic conditions (Adams et al., 2012). In conventional aquaculture systems, however, signicant concentrations of CH4 are typically unlikely (Rakocy et al., 2006). N2O is an important GHG which has a global warming potential (GWP) that is 296 times higher that of CO2 on a 100-year timescale (IPCC, 2007). Hu et al. (2012) estimated that aquaculture could contribute up to 5.72% of global anthropogenic N2ON emission by 2030 if it continues to increase at the present annual growth rate. However, the mechanisms for N2O generation in an aquaculture system are highly inuenced by environmental conditions and are not well known. Both nitrication and denitrication can contribute to the emission of N2O from aquaculture systems (Beaulieu et al., 2011; Hu et al., 2013a). The addition of carbohydrate to aquaculture systems could decrease the abundance of nitrifying bacteria, and thus may minimize nitrication-driven N2O production (Avnimelech, 1999). Furthermore, the supplementation of external organic carbon could provide electron donors for denitrication and may reduce N2O generation through denitrication (Hu et al., 2013b). However, carbohydrate addition might increase the concentration of organic materials, which may lead to higher CO2 and CH4 emissions. Nonetheless, there has been a dearth of studies thus far on the emissions of GHGs from zero-water exchange intensive aquaculture system. The purpose of this study was to investigate the effect of soluble starch addition on GHG emissions from a zero-water exchange tilapia intensive aquaculture system. Nitrogen transformations in aquaculture system with and without soluble starch addition were also examined.
2. Materials and methods 2.1. System setup Two aquaculture systems were placed side by side in an airconditioned room at a temperature around 25 C and exposed to 24-h lighting. The schematic diagram of aquaculture system is shown in Fig. 1. The system was mainly composed of a plastic tank (KMT85, Tuff Stuff, Terra Bella, CA, USA), with a working volume of 200 L. Air was supplied continuously by an air pump through three diffusers placed at the bottom of the tank, and desired DO concentrations were obtained by controlling the air ow rate. Aeration also provided mixing for the system. In each tank, one biolter composed of mesh nylon biolter media bags lled with 1.5 kg of biomedia (Kaldnes @ media, Aquatic Eco-System, Apopka, FL, USA), was placed adjacent to the air diffusers to facilitate the growth of nitrifying bacteria. Semi-transparent acrylic plastic lids were used to minimize water loss by evaporation and to prevent algal growth. Freshwater was occasionally added to the system to compensate water loss through evaporation, when necessary. The aquaculture tank was maintained at pH of around 7.0 by periodic dosing of sodium bicarbonate (NaHCO3). 2.2. Experiment design It generally requires about 4 weeks to establish the required microbial community in a biolter of an aquaculture system (Avnimelech, 2009). However, since heterotrophic bacteria typically have a maximum growth rate ve-fold faster and biomass yields two to three-fold greater than that of nitrifying bacteria, bioocs could be established rapidly in a matter of days (Grady and Lim, 1980). To overcome the time difference, efuent from a stable operating aquaculture system was used as the initial tank water for the experiment, and the biolters were inoculated in the stable operating aquaculture system for two weeks prior to the start of the experiment. This facilitated the establishment of nitrifying bacteria in the biolters, which were used in subsequent experiments. Each tank was stocked with mixed sex tilapia sh (Oreochromis niloticus) with an average weight of 120.5 27.3 g, to obtain an initial stocking density of around 23.5 kg/m3. The sh were obtained from Windward Community College (Honolulu, Hawaii, USA). Fish were grown without water exchange for 8 weeks, and were fed once daily at 10:00 A.M. with 42% protein commercial aquatic feed pellets (Silver Cup Trout Feed, Tooele, UT, USA). The amount of feed per feeding time was determined based on sh response to previous feeding (CasillasHernndez et al., 2006). Ten minutes after feeding, all the feed pellets remaining above water surface were collected, dried and weighed. The feeding rate was adjusted in the subsequent days so that the leftover (un-consumed) feed 10 min after feeding was no more than 5% of the total feed added. In the treatment tank, soluble starch was added daily along with feed to maintain a C/N ratio (w/w) of 16:1; while the control tank was supplied with sh feed only (Nootong et al., 2011). The amount of daily starch addition was calculated as per the equations proposed by Avnimelech (1999), and about 1.4 g of soluble starch was added for each gram of formulated sh feed. The pre-weighed soluble starch was completely mixed with tank water in a beaker and was then uniformly sprayed over the tank surface water after each feeding. 2.3. Analytical method Water samples from each tank were obtained after feeding every other day and were analyzed immediately for TAN, NO 2 , NO3 , total phosphate (TP), and chemical oxygen demand (COD) concentrations using HACH reaction kits (Loveland, CO, USA), namely Ammonia TNTplus (TNT 830), Nitrite TNTplus (TNT 839), Nitrate TNTplus (TNT 836), Phosphorus TNTplus (TNT 845), and COD Reagent TNTplus (TNT 822), respectively. Dissolved oxygen (DO) concentrations, temperature,
195
pH, and salinity were measured in situ daily using the HQ40d Portable Water Quality Lab Package (HACH, Loveland, CO, USA). Bioocs concentration in water, expressed as total suspended solids (TSS), was determined according to the Standard Methods (APHA, 2005). The bioocs attached to the biomedia of the biolter were washed out with pure water at the end of the experiment, ltered, dried at 105 C and weighed. A LECO TruSpec CN analyzer (LECO Corp., St. Joseph, MI, USA) was used to measure the total nitrogen (TN) contents of sh mass, sh feed and bioocs. Diurnal variations of GHG emission rates were monitored once a week during the last three weeks of the study period, after attaining fairly constant daily feed consumption. Dissolved N2O concentrations in the aqueous phase were determined using a Clark type electrode N2O sensor (Unisense, Aarhus N, Denmark) (Andersen et al., 2001). N2O concentrations in the gas phase were calculated based on the following equation (Rassamee et al., 2011), CN2 O;air kL a SN2 O V r =Q air V r kL a=H 1
sh growth performance were found between the treatment and control tanks. Nearly the same average sh weight increase was recorded in both tanks. However, the use of mixed sex tilapia as the test subject resulted in unexpected breeding during the study period, particularly in the treatment tank. The larvae were collected with a ne mesh net and weighed after counting. During the study period, about 246 and 37 tilapia larvae were collected from treatment and control tanks, respectively. The wet weight of the collected larvae in treatment and control tanks were 19.6 g and 3.5 g, respectively. The same feed conversion ratio (FCR) of 2.2 was obtained in both treatment and control tanks. 3.2. Water quality variation The ranges of pH, DO concentrations, and temperature in the treatment and control tanks are presented in Table 2. All of the parameters were similar and were not suspected to be growth limiting. It is worth noting that the aeration rates of treatment and control tanks were different. The addition of soluble starch to the treatment tank resulted in a decrease of DO concentrations due to the enhanced growth of heterotrophic bacteria. In order to maintain a comparable DO concentration in both tanks, the control tank was supplied with 4.7 L/min of air, whereas the aeration rate in the treatment tank was gradually increased from 5.1 to 9.8 L/min to compensate for increasing oxygen consumption resulting from higher bioocs concentration in the tank. The variations of TAN, NO 2 , and NO3 concentrations during the study period are presented in Fig. 2. The variations of TSS, COD, and TP concentrations in the treatment and control tanks are presented in Fig. 3. Accumulation of TAN was observed following sh stocking in both tanks. The highest TAN concentrations of 6.78 mg N/L and 2.84 mg N/L were observed on day 26 and day 22 in treatment and control tanks, respectively. In addition, TAN accumulation lasted for a longer period of time in the treatment tank. It required 36 and 28 days for TAN concentrations to decrease to less than 1.0 mg N/L in the treatment and control tanks, respectively. The TAN concentrations in the treatment tank were, however, consistently lower than that in the control tank at the later stage of the study period. Signicant increases in NO 2 and decreases in NO3 concentrations were observed in the treatment tank after the start of the experiment.
Table 1 Growth performances of sh in control and treatment tanks. Parameters Average initial weight (g) (n = 40) Average nal weight (g) (n = 40) Fish biomass increasea (g) Fish feed consumption (g) Feed conversion ratio
a
where CN2 O;air is the gas phase N2O concentration (mol/L); kLa is the volumetric mass transfer coefcient (1 / h), which varied with different air ow rates and is determined according to equation provided by Foley et al. (2010); SN2 O is the concentration of N2O in the liquid phase (mol/L); Vr is the working volume of the tank (L); Qair is the air ow rate (L/h); H is the Henry's constant, which is 1.303 for N2O at 25 C (Sander, 1999). The gas phase concentrations of CO2 and CH4 were determined by using a Varian CP-4900 micro gas chromatograph (Varian Inc., Walnut Creek, CA, USA), equipped with thermal conductivity detector and a 10 m PPQH BF column. Helium maintained at 30 mL/min was used as a carrier gas, and the temperature of the column and injector were set at 80 C and 100 C, respectively. 2.4. Statistical analysis Experimental data were presented as a mean of three replicates. Statistical analyses were performed using SPSS v16.0 for Windows software (SPSS v16.0 IBM Corporation, Somers, NY, USA). The signicance for all p-values was 0.05 unless otherwise stated. 3. Results 3.1. Growth performance of tilapia The growth performances of tilapia in both tanks are shown in Table 1. During the 8-week study period, no sh mortality was observed in both treatment and control tanks. Also, no signicant differences in
Includes the weight of tilapia larvae collected during the research period.
196
Table 2 Physical parameters of water in treatment and control tanks. Parameters pH DO (mg/L) Temperature (C) Treatment (n = 56) 7.08 0.07 4.49 0.33 24.6 0.3 Control (n = 56) 7.04 0.06 4.53 0.38 24.4 0.3
Both NO 2 increases and NO3 decreases continued until day 14. The highest NO2 concentration of 7.2 mg N/L was observed on day 14 in the treatment tank. The NO 2 concentration then decreased rapidly to less than 0.4 mg N/L after day 22. In the control tank, a relatively constant accumulation of NO 2 and a linear increase in NO3 concentrations were observed. Similar to TAN proles, the treatment tank showed lower NO 2 and NO3 concentrations during the later stage of the study period. The TSS concentrations in the treatment tank increased from 52.5 7.3 mg/L to 796.5 18.8 mg/L during the study period, while it remained fairly constant at around 47.5 9.2 mg/L in the control tank. In addition, although the nitrogen contents of bioocs in both tanks were initially around 6.1 0.2%, the nitrogen content in the treatment tank decreased to 5.3 0.3%, but remained fairly constant in the control tank following the 8-weeks study period. The COD concentration increased at a much faster rate in the treatment tank than in the control tank. At the end of the study period, the COD concentration in the treatment tank was 2.8-fold higher than that in the control tank. Statistical analysis showed that starch addition had no signicant effect on TP removal. Similar TP concentration proles were found in both the treatment and control tanks.
Fig. 3. Variation of total suspended solids (A), chemical oxygen demand (B), and total phosphate (C) concentrations in treatment and control tanks during the study period. Values are mean of triplicate samples analyzed for each parameter.
3.3. Daily GHG emissions Fig. 4 shows diurnal variations of CO2 and N2O emission rates in the treatment and control tanks. The time-weighted GHG emissions from treatment and control tanks are listed in Table 3. CH4 concentrations were consistently below the detection limits during the study period (data not shown), and was not included in the calculation of GHG
emissions in this study. The treatment tank had higher CO2 emission rates and lower N2O emission rates than the control tank. The addition of soluble starch increased the daily CO2 emission by 91.1%, and reduced the daily N2O emission by 83.4%. Slight increases in the CO2 emission rate was observed after feeding in both tanks.
Fig. 2. Variation of total ammonia nitrogen (A), nitrite (B), and nitrate (C) concentrations in treatment and control tanks during the study period. Values are mean of triplicate samples analyzed for each parameter.
Fig. 4. Diurnal variation of CO2 (A) and N2O (B) emission rates in treatment and control tanks. Each data point is the mean of three different measurements on different days.
Z. Hu et al. / Science of the Total Environment 470471 (2014) 193200 Table 3 Greenhouse gases emissions and N2ON conversion ratio in treatment and control tanks. Parameters CO2 emission (g/d) (n = 3) N2ON emission (mg N/d) (n = 3) Feed nitrogen input (mg N/d) (n = 3) N2O conversion ratio (%)a
a
197
To investigate the major factors inuencing GHG emissions, the di urnal variation of TAN, NO 2 , NO3 , and COD concentrations in both treatment and control tanks were also determined (Fig. 5). The control tank showed higher TAN, NO 2 , and NO3 concentrations than the treatment tank, whereas higher COD concentrations were observed in the treatment tank. All of the water quality parameters were relatively stable within a short time span of 24 h. 4. Discussion 4.1. Effect of soluble starch addition on sh performance In order to minimize inorganic nitrogen accumulation and to accelerate bioocs formation, organic carbon can be added either as an external organic carbon source or by changing the feed composition to increase its organic carbon content (Avnimelech, 1999). Both approaches were reported in the literature without identifying their inherent merits (Azim and Little, 2008; Asaduzzaman et al., 2010). In
this study, we used the former approach because it is easier to implement. Adouani et al. (2001) reported that compared with shorter chain carbon compounds (e.g., ethanol, acetate), long chain carbon compounds (e.g., peptone, starch) could produce less N2O when used as the carbon source for denitrication, mainly because of its lower electron supply rate. Soluble starch is a cheap and widely used carbon source and was thus employed in this study. Hargreaves (1998) reported that about 11 to 36% of the nitrogen in sh feed could be recovered in an aquaculture system. In this study, tilapia in both the treatment and control tanks exhibited the same feed conversion ratio (FCR) of 2.2, meaning that about 18% of the nitrogen in the feed was recovered as sh mass. Interestingly, no signicant difference was found in the FCR between treatment and control tanks. Studies reported that the addition of carbohydrates resulted in the production of microbial proteins that could serve as sh food, thereby increasing the FCR (Avnimelech, 1999; Crab et al., 2012). In this study, signicant increase in reproductive activity was observed in the treatment tank as evidenced from the spawning of sh. During the breeding process, most of the energy obtained from feeding was reported to be utilized in gonad development (Widanarn et al., 2012). Thus, the availability of additional feed through bioocs formation might be compensated by their enhanced consumption during breeding. Several studies reported the enhancement effect of bioocs on the reproduction of tilapia and shrimp (Emerenciano et al., 2011; Widanarn et al., 2012). It is also important to note that the high TSS concentrations in the treatment tank may also have a negative impact on sh growth. Severe growth inhibition and increased mortality of tilapia were reported when the suspended solids concentration exceeded 850 mg/L (Little
Fig. 5. Diurnal variations of total ammonia nitrogen (A), nitrite (B), nitrate (C), and chemical oxygen demand (D) concentrations in treatment and control tanks. Each data point is the mean of three different measurements on different days.
198
et al., 2008). It is recommended that the maximum suspended solids concentration in an aquaculture system should not exceed 500 mg/L (Azim and Little, 2008; Little et al., 2008). This could be another reason for the same FCR in both treatment and control tanks. The integration of a clarier with the aquaculture tank may help to maintain a low TSS concentration in aquaculture tank. Daily addition of soluble starch inuenced the formation of bioocs, as shown in Fig. 3A. During the 8-week study period, the TSS concentration (representing bioocs) in the treatment tank increased by 14-folds. The TSS concentration in the control tank, however, remained relatively constant. The accumulation of bioocs in the treatment tank was attributed to the enhancement effect of added starch on the growth of heterotrophic bacteria, and the uptake of bioocs by sh was apparently not sufcient to prevent its buildup. This nding was in close agreement with that of Nootong et al. (2011), who reported an increase in TSS concentration from 52 to 1180 mg/L in their tilapia aquaculture tank that received feed and tapioca starch, while the TSS concentration remained relatively constant at 99 mg/L in control tanks which received feed only. Based on statistical analyses, the nitrogen contents of bioocs in the treatment and control tanks were signicantly different at the end of the study period. Bioocs in the treatment tank showed lower nitrogen contents than that in the control tank. This is mainly attributed to the accumulation of soluble microbial products (SMP) and unconsumed starch. During substrate metabolism, bacterial growth, and bacterial decay, SMP is produced (Barker and Stuckey, 1999). The production rate of SMP is proportional to the concentration of bacteria and their retention time (or age) (Jarusutthirak and Amy, 2006). Since there were no water exchange during the study period and high concentration of bioocs was observed in the treatment tank, it is reasonable to believe that high accumulation of SMP occurred in the treatment tank. Furthermore, soluble starch was added to the treatment tank according to the theoretical quantity calculated based on equations reported by Avnimelech (1999). However, the added starch may not have been completely utilized for microbial metabolism. Gao et al. (2012) reported that the optimum carbohydrate addition was 75% of its theoretical quantity in a zerowater exchange system cultured with Whiteleg shrimp (Litopenaeus vannamei). The accumulation of SMP and starch was evident from the higher COD concentrations in the treatment tank (Fig. 3B). Bioocs is composed of micro-organisms and a mixture of non-living detritus materials. Micro-organisms only account for 220% of the organic fraction of bioocs (Wilen et al., 2003). Part of SMP and unconsumed starch could be absorbed by bioocs, thus reducing its nitrogen content.
Although the DO concentrations in both tanks were maintained above 4.0 mg/L throughout the study period, the existence of microanoxic zones was likely in the inner part of the bioocs where denitrication could have occurred (Zeng et al., 2003). In the treatment tank, the presence of nitrates and the addition of soluble starch stimulated denitrication during the initial stages of the study period. High nitrate reduction rates were observed, which also led to an internal accumulation of nitrite (Fig. 2). Relatively low concentrations of nitrate and nitrite were achieved in the treatment tank three weeks after sh stocking, which was most likely associated with denitrication. Nitrogen mass balances of the treatment and control tanks throughout the study period were conducted and the results are presented in Fig. 6. The composition of nitrogen input for both treatment and control tanks were similar. Fish feed contributed to over 80% of the nitrogen input, and about 15% of the nitrogen inputs were from inorganic nitrogen species, which were already present in the tank water before the start of the experiment. A FCR of 2.2 was obtained in both tanks, indicating that about 14% of the nitrogen output was associated with sh mass increase. In the treatment tank, only about 0.4% of the nitrogen output was distributed as inorganic nitrogen species in the tank water, while it accounted for as high as 48.5% of the nitrogen output in the control tank. There was also a signicant difference in unaccounted nitrogen (mainly gaseous nitrogen loss, including N2, N2O and NH3) between the treatment and control tanks. Since the pH in both tanks was around 7.0, and the major fraction of TAN was in ionized form (i.e., NH+ 4 ), ammonia volatilization was likely to be negligible. The N2O emissions during the study period were also considered to be insignicant compared to N2 (as discussed in Section 4.3). Thus, N2 produced through denitrication was considered to be the major source of unaccounted nitrogen. Gaseous nitrogen loss in the treatment tank accounted for about 76.2% of the nitrogen output. This is because the addition of soluble starch provided sufcient organic carbon for denitrication. The nitrogen loss in this study was much higher than that of Nootong et al. (2011) who reported a nitrogen gas loss of 32% in zero-water exchange tilapia cultivation tanks supplied with tapioca starch at a C/N weight ratio of 16:1. This could be attributed to a high initial NO 3 concentration in the tank water at the beginning of the experiment. On the other hand, about 33.3% of nitrogen output was distributed as gaseous nitrogen loss in the control tank. This was similar to the results of Thakur and Lin (2003) who reported gaseous nitrogen loss of 36% in their closed aquaculture system without carbohydrate addition.
4.2. Effect of soluble starch addition on nitrogen transformations After sh stocking, TAN concentration started to accumulate followed by decrease and remained fairly constant at the later stages in both the treatment and control tanks (Fig. 2A). However, the treatment tank showed higher peak of TAN concentrations and required a longer period of time to decrease to a relatively stable TAN concentration. This was caused by the addition of soluble starch. In aquaculture systems, most of the TAN produced by the metabolism of sh are removed through nitrication. During the initial stages of the study period, there was not enough nitrifying bacteria in the system to remove all of the TAN excreted by sh, and part of the TAN was accumulated in the system. However, the TAN concentration decreased slowly and was able to reach low stable concentrations with the growth of nitrifying bacteria at the later stages. In the treatment tank, while the addition of soluble starch stimulated the growth of heterotrophic bacteria, it inhibited the growth of nitrifying bacteria, due to competition for nutrients (especially nitrogen) between heterotrophic and nitrifying bacteria. Michaud et al. (2006) also reported the inhibition effect of organic carbon on nitrifying bacteria in aquaculture systems. The lower growth rate of nitrifying bacteria in the treatment tank led to higher TAN accumulation and more time was needed to attain stable concentrations.
199
4.3. Effect of soluble starch addition on GHG emissions Soluble starch addition signicantly increased the daily CO2 emissions from the treatment tank. This was mainly due to the following reasons: (i) starch addition stimulated the growth of heterotrophic bacteria, which could convert nearly 50% of the organic matter (stoichiometrically) into CO2 under non-limiting oxygen conditions; (ii) part of the starch added to the treatment tank was unconsumed and could be absorbed by bioocs (the unconsumed starch was converted to CO2 by heterotrophic bacteria); and (iii) higher aeration rates in the treatment tank (to maintain similar DO concentrations with the control tank) led to higher stripping effects. On the other hand, the addition of soluble starch reduced the N2O emissions from the treatment tank. In aquaculture systems, both nitrication and denitrication are responsible for the emission of N2O (Hu et al., 2013b). The abundance and activity of nitrifying bacteria in the treatment tank was inhibited by the addition of soluble starch, which was evident from the higher TAN accumulation and lower TAN removal rate (Fig. 2A). With no signicant differences in operational parameters (i.e., DO, pH and temperature) and TAN concentrations, less nitrifying bacteria led to lower nitrication-driven N2O production. Furthermore, the addition of soluble starch could have stimulated the occurrence of denitrication, which was apparent from the high nitrate reduction rate in the treatment tank (Fig. 2C). N2O is an important intermediate during denitrication (Hu et al., 2012). However, the addition of soluble starch provided enough electron donors for denitrication, thus avoiding the accumulation of N2O and consequently reducing N2O emissions from denitrication (Hu et al., 2013b). It is worth noting that at the end of the experiment, the control tank water contained much higher NO 3 concentrations than the treatment tank water. The NO 3 in the efuent may contribute to the N2O emissions of the surrounding environments (Wang et al., 2009). Soluble starch reduced the daily N2O emissions by 5.7 g CO2,eq, while it increased the daily CO2 emissions by 28.9 g, with respect to the control tank. Thus, the addition of soluble starch actually increased the daily GHG emissions from the aquaculture system by 23.2 g CO2,eq, or as much as 60.2% more than the control tank. In addition, higher aeration was needed for the treatment tank to maintain the DO concentrations similar to the control. At the end of the study period, the aeration rate for the treatment tank was more than twice that for the control tank. Taking higher energy consumption for aeration into consideration, soluble starch addition had an overall adverse effect in controlling GHG emissions from the aquaculture system. Although the same FCR was obtained in treatment and control tanks in this study, several studies reported that the uptake of bioocs by aquatic animals could improve their production performance (Avnimelech, 2007; Kuhn et al., 2009; Nootong et al., 2011). Further studies using monosex species and lower TSS concentrations are necessary to closely determine the economic and environmental impacts of carbohydrate addition on minimizing GHG emissions from aquaculture systems. 5. Conclusions Soluble starch addition had signicant effects on the nitrogen transformations and GHG emissions in a zero-water exchange intensive aquacul ture system. Lower TAN, NO 2 , and NO3 concentrations were obtained in an aquaculture system with soluble starch addition, which was attributed to the enhancement of heterotrophic bacterial growth and denitrication. The addition of soluble starch enhanced denitrication and about 76.2% of the nitrogen input was emitted as gaseous nitrogen (i.e., N2 and N2O) in the treatment tank; which was much higher than what was observed in the control tank (33.3%). Although soluble starch addition minimized N2O emissions from the aquaculture system, it resulted in signicantly higher CO2 emissions. Overall, the addition of soluble starch increased the daily GHG emissions (as CO2 equivalent) from the aquaculture system
by 60.2%, indicating that it may have signicant adverse effects in controlling GHG emissions from commercial aquaculture systems. Conict of interest We declare that we have no nancial and personal relationships with other people or organizations that can inappropriately inuence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as inuencing the position presented in, or the review of, the manuscript entitled Inuence of carbohydrate addition on nitrogen transformations and greenhouse gas emissions of intensive aquaculture system. Acknowledgments This work is being supported by National Research Foundation Grant funded by the Korean Government (NRF-2011-220-D00071), National Natural Science Foundation of China (No. 21307076), and Supplemental Research and Extension Grant from the College of Tropical Agriculture and Human Resources (CTAHR), University of Hawaii at Manoa (UHM), USA. We would like to thank Dr. Clyde Tamaru for assistance with experiment design and Dr. Devin Takara for his comments on the manuscript. References
Adams CA, Andrews JE, Jickells T. Nitrous oxide and methane uxes vs. carbon, nitrogen and phosphorous burial in new intertidal and saltmarsh sediments. Sci Total Environ 2012;434:24051. Adouani N, Lendormi T, Limousy L, Sire O. Effect of the carbon source on N2O emissions during biological denitrication. Resour Conserv Recycl 2001;54:299302. Andersen K, Kjr T, Revsbech NP. An oxygen insensitive microsensor for nitrous oxide. Sensors Actuators B Chem 2001;81:428. APHA. Standard methods for the examination of water and wastewater. 21th ed. Washington, DC: APHA; 2005. Asaduzzaman M, Rahman MM, Azim ME, Islam MA, Wahab MA, Verdegem MC, et al. Effects of C/N ratio and substrate addition on natural food communities in freshwater prawn monoculture ponds. Aquaculture 2010;306:12736. Avnimelech Y. Carbon/nitrogen ratio as a control element in aquaculture systems. Aquaculture 1999;176:22735. Avnimelech Y. Bio-lters: the need for a new comprehensive approach. Aquacult Eng 2006;34:1728. Avnimelech Y. Feeding with microbial ocs by tilapia in minimal discharge bio-ocs technology ponds. Aquaculture 2007;264:1407. Avnimelech Y. Biooc technology a practical guide book. Baton Rouge, Louisiana: The World Aquaculture Society; 2009. Azim ME, Little DC. The biooc technology (BFT) in indoor tanks: water quality, biooc composition, and growth and welfare of Nile tilapia (Oreochromis niloticus). Aquaculture 2008;283:2935. Barker DJ, Stuckey DC. A review of soluble microbial products (SMP) in wastewater treatment systems. Water Res 1999;33:306382. Beaulieu JJ, Tank JL, Hamilton SK, Wollheim WM, Hall RO, Mulholland PJ, et al. Nitrous oxide emission from denitrication in stream and river networks. PNAS 2011;108:2149. Casillas-Hernndez R, Magalln-Barajas F, Portillo-Clarck G, Pez-Osuna F. Nutrient mass balances in semi-intensive shrimp ponds from Sonora, Mexico using two feeding strategies: trays and mechanical dispersal. Aquaculture 2006;258:28998. Crab R, Avnimelech Y, Defoirdt T, Bossier P, Verstraete W. Nitrogen removal techniques in aquaculture for a sustainable production. Aquaculture 2007;270:114. Crab R, Defoirdt T, Bossier P, Verstraete W. Biooc technology in aquaculture: benecial effects and future challenges. Aquaculture 2012;356357:3516. Delong DP, Losordo TM, Rakocy JE. Tank culture of tilapia. Auburn, USA: SRAC Publication; 2009. Durborow RM, Crosby DM, Brunson MW. Nitrite in sh ponds. Auburn, USA: SRAC Publication; 1997. Emerenciano M, Cuzon G, Goguenheim J, Gaxiola G. Floc contribution on spawning performance of blue shrimp Litopenaeus stylirostris. Aquacult Res 2011;44:111. FAO. The state of world sheries and aquaculture, 2012. Rome, Italy: Food and Agricultural Organization of the United Nations; 2012. Foley J, de Haas D, Yuan Z, Lant P. Nitrous oxide generation in full-scale biological nutrient removal wastewater treatment plants. Water Res 2010;44:83144. Gao L, Shan HW, Zhang TW, Bao WY, Ma S. Effects of carbohydrate addition on Litopenaeus vannamei intensive culture in a zero-water exchange system. Aquaculture 2012;342343:8996. Good C, Davidson J, Welsh C, Snekvik K, Summerfelt S. The effects of carbon dioxide on performance and histopathology of rainbow trout Oncorhynchus mykiss in water recirculation aquaculture systems. Aquacult Eng 2010;42:516. Grady CPL, Lim HC. Biological wastewater treatment: theory and applications. New York, USA: Marcel Dekker; 1980.
200
Z. Hu et al. / Science of the Total Environment 470471 (2014) 193200 Nootong K, Pavasant P, Powtongsook S. Effects of organic carbon addition in controlling inorganic nitrogen concentrations in a biooc system. J World Aquacult Soc 2011;42:33946. Pillay TVR, Kutty MN. Aquaculture: principles and practices. 2nd ed. Oxford, UK: Blackwell Publishing Ltd.; 2005. Rakocy JE, Masser MP, Losordo TM. Recirculating aquaculture tank production systems aquaponics integration sh and plant culture. Auburn, USA: SRAC Publication; 2006. Rassamee V, Sattayatewa C, Pagilla K, Chandran K. Effect of oxic and anoxic conditions on nitrous oxide emissions from nitrication and denitrication processes. Biotechnol Bioeng 2011;108:203645. Sander R. Compilation of Henry's Law constants for inorganic and organic species of potential importance in environmental chemistry (version 3). http://www.mpchmainz.mpg.de/~sander/res/henry.html, 1999. Thakur DP, Lin CK. Water quality and nutrient budget in closed shrimp (Penaeus monodon) culture systems. Aquacult Eng 2003;27:15976. Tomasso JR. Toxicity of nitrogenous wastes to aquaculture animals. Rev Fish Sci 1994;2: 291314. Wang S, Liu C, Yeager KM, Wan G, Li J, Tao F, et al. The spatial distribution and emission of nitrous oxide (N2O) in a large eutrophic lake in eastern China: anthropogenic effects. Sci Total Environ 2009;407:33307. Widanarn Ekasari J, Maryam S, et al. Evaluation of biooc technology application on water quality and production performance of red tilapia Oreochromis sp. cultured at different stocking densities. HAYATI 2012;19:738. Wilen BM, Jin B, Lant P. The inuence of key chemical constituents in activated sludge on surface and occulating properties. Water Res 2003;37:212739. Williams J, Crutzen PJ. Nitrous oxide from aquaculture. Nat Geosci 2010;3:143. Zeng RJ, Lemaire R, Yuan Z, Keller J. Simultaneous nitrication, denitrication, and phosphorus removal in a lab-scale sequencing batch reactor. Biotechnol Bioeng 2003;84:1708.
Hargreaves JA. Nitrogen biogeochemistry of aquaculture ponds. Aquaculture 1998;166: 181212. Hargreaves JA, Tucker CS. Managing ammonia in sh ponds. Auburn, USA: SRAC Publication; 2004. Hu Z, Lee JW, Chandran K, Kim S, Khanal SK. Nitrous oxide (N2O) emission from aquaculture system: a review. Environ Sci Technol 2012;46:647080. Hu Z, Lee JW, Chandran K, Kim S, Sharma K, Brotto AC, et al. Nitrogen transformations in intensive aquaculture system and its implication to climate change through nitrous oxide emission. Bioresour Technol 2013a;130:31420. Hu Z, Zhang J, Xie H, Liang S, Li S. Minimization of nitrous oxide emission from anoxic oxic biological nitrogen removal process: effect of inuent COD/NH+ 4 ratio and feeding strategy. J Biosci Bioeng 2013b;115:2728. IPCC. Changes in atmospheric constituents and in radioactive forcing. In: Solomon S, et al, editor. Climate change 2007: The physical science basis. Contribution of working group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press; 2007. p. 11443. Jarusutthirak C, Amy G. Role of soluble microbial products (SMP) in membrane fouling and ux decline. Environ Sci Technol 2006;40:96974. Kuhn DD, Boardman GD, Lawrence AL, Marsh L, Flick JGJ. Microbial oc meal as a replacement ingredient for sh meal and soybean protein in shrimp feed. Aquaculture 2009;296:517. Little DC, Murray FJ, Azim E, Leschen W, Boyd K, Young JA, et al. Options for producing a warm-water sh in the UK: limits to Green Growth? Trends Food Sci Technol 2008;19:25564. Michaud L, Blancheton JP, Bruni V, Piedrahita R. Effect of particulate organic carbon on heterotrophic bacterial populations and nitrication efciency in biological lters. Aquacult Eng 2006;34:22433.