- Email: [email protected]
The reduction of Chlorella vulgaris concentrations through UV-C radiation treatments: A nature-based solution (NBS)
Environmental Research 156 (2017) 183–189
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
The reduction of Chlorella vulgaris concentrations through UV-C radiation treatments: A nature-based solution (NBS) Erika S. Chena, Thomas B. Bridgemanb, a b
MARK
⁎
Sylvania Northview High School, Sylvania, OH 43560, USA Lake Erie Center, University of Toledo, Oregon, OH 43616, USA
A R T I C L E I N F O
A BS T RAC T
Keywords: Algal concentration UV radiation HAB Mortality Density-dependent NBS Modeling
Algal blooms have become a pressing issue in inland freshwater systems on local and global scales. A plausible approach to reducing algae without the use of chemical/biological agents is through the use of UV-C radiation from lamps potentially powered by in situ solar panels to eliminate algae. Yet, the quantitative scientific base has not been established. Our objective is to conduct a controlled experiment to quantify the effectiveness of UV-C radiation on the reduction of Chlorella vulgaris, a common algal species in the Great Lakes region. A full factorial design of three intensities of UV-C radiation (0, 15, and 30 W) and three sources of C. vulgaris was constructed to test the corresponding hypotheses. Empirical models were constructed to predict the reductions. UV-C radiation effectively reduced the algal concentration with clear differences by radiation level and source of algal water. Algal concentration decreased exponentially over time, with distinct decreasing trends among the radiation intensities and the samples. With 15 W UV-C radiation, algal concentration of three samples were reduced to 75.3%, 51.5%, and 70.0% of the initial level within an hour, respectively. We also found a clear density-dependent reduction rate by UV radiation. Using this information, more efficient treatment systems could be constructed and implemented for cleaning algae-contaminated water.
1. Introduction Algal blooms have become a major environmental issue in many inland lakes (Pogge, 2015) and are major concerns of the general public at a regional, national, and global level (Hallegraeff, 1993; Ho and Michalak, 2015). In the Midwest region of the United States, Lake Erie has had a history of harmful algal blooms (HABs), with a maximum level in the summer of 2015 (cf Ouyang et al., 2017). Another algal bloom event of a very similar magnitude in Lake Erie erupted in 2011 – the largest algal bloom in the lake's history up to that point, which had a peak over three times greater than all previous algal blooms. Like the 2015 algal bloom, the main cause was eutrophication. Intensified agricultural practices and meteorological conditions in the spring were directly related with increased phosphorus depositions in the western basin of the lake (Chaffin et al., 2014). Other factors contributing to the issue were the extended period of weak lake circulation, which prevented the flushing of nutrients from the system (Michalak et al., 2013). Although not as large as the HABs of 2011 and 2015, the 2014 algal bloom still had a great impact on the general public. More than 400,000 people were without safe drinking water for 2.5 days during the 2014 blooms (Jetoo et al., 2015).
⁎
Corresponding author. E-mail address: [email protected] (T.B. Bridgeman).
http://dx.doi.org/10.1016/j.envres.2017.03.007 Received 2 January 2017; Received in revised form 2 March 2017; Accepted 4 March 2017 0013-9351/ © 2017 Elsevier Inc. All rights reserved.
HABs are not restricted to the Great Lakes. A similar case was found in the Southeastern United States. From 1993 to 2000, there were observations of potential microcystin-producing cyanobacteria in 74% of streams in Alabama, Georgia, South Carolina, and North Carolina (Loftin et al., 2016). In the North Sea, HABs have had many negative impacts, including poisoning, death, economic damage to aquaculture, and decline in tourism. The area has observed a strong increase in algal blooms since the Industrial Revolution. The main causes of the algal blooms were ballast water dispersal, eutrophication, and global warming. However, nutrients were not the biggest factor in determining the occurrence of HABs. Instead, the main factors were the initial density and interspecific competition (Coppen, 2015). Another key factor affecting algae growth is the supply of carbon dioxide (CO2). In a study quantifying the relationship of the photosynthesis of algae (Anabaena variabilis) to CO2 concentration, the author indicated that as the CO2 concentration increased, the photosynthetic properties of algae were greater (Kaplan et al., 1980). Consequently, developing environmentally safe and economically affordable methods and systems has become a challenge in cleaning up contaminated freshwater (Heisler et al., 2008). Recently developed concepts and principles of nature-based solutions (NBS) (Maes and
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
water depth will be elevated as the surface water gets more transparent due to reduction of algae concentration near the surface. Consequently, the total algal mortality in the entire water column will not be maintained over time, but will decrease in effectiveness, resulting in exponential changes in mortality (i.e., H1). Such a change is also dependent of the intensity of UV-C radiation, with a higher radiation causing more mortality. The effectiveness of UV-C radiation may also vary by the source of algae if the algae were cultivated using different lights and media (i.e., their resistances to UV-C radiation are not the same). Finally, one would expect that the same level of UV-C radiation will cause more algae to die when the algal concentration is high (H2). Consequently, the relative reduction in algal concentration would be higher for water with high algal concentration than that with low concentration (H2).
Jacobs, 2015) appear plausible and stimulating for the scientific community in seeking a NBS for treating polluted water. Based on this concept, we sought an approach of applying UV that can be created in the field using solar panels on algae-polluted water for reducing algal abundance. Our long-term study objective is to develop a scientifically-sound, portable and user-friendly system that can be used to treat algaecontaminated water. A NBS approach to the matter is through the use of UV radiation. Germicidal UV (UV-C) radiation has been proven to possess the ability to eliminate algae and other microorganisms. UV-C radiation is able to eliminate such microorganisms by deactivating the DNA and damaging the nucleic acids. UV-C radiation has not only been effective in treatment, but is also inexpensive, low maintenance, and simple. In 1983, a design for a water purifying tank was created that incorporated UV-C radiation and biological oxidation. The UV-C radiation was able to inhibit the growth of bacteria (Takeguchi and Oyobe, 1983). UV-C radiation has also been used in disinfecting hospitals. In a 13-year study, a UV apparatus was placed in the water main of a new hospital building. Before the construction of the new hospital, 27% of water samples taken from the old hospital contained Legionella. After 13 years, the UV-C radiation usage resulted in negative water cultures and no cases of nosocomial Legionella pneumophila (Hall et al., 2003). This study demonstrated the effectiveness of UV-C radiation. Along with microorganisms such as bacteria and mold, UV-C radiation has also been proven to eliminate harmful algal species. For example, when exposed to UV-C treatments for 21 days, treatments of Chlorella minutissima experienced chlorophyll bleaching, oxidative stress, programmed cell death, and inhibited metabolic activity (Borderie et al., 2014). UV-C is also capable of reducing microcystin-LR (MC-LR) release risk of Microcystis aeruginosa, the primary cause of Lake Erie HABs (Ou et al., 2011). As UV-C dosages increased, the photosynthetic capacities of M. aeruginosa were reduced, accompanied by a complete degradation of MC-LR immediately after irradiation. Unfortunately, quantifying the applications of UV-C radiation to Chlorella vulgaris has not been studied extensively. Our literature search did not yield any publications that have the scientific base for determining the treatment duration for common algal species in our freshwater lakes and rivers in the Great Lakes region. Our objective is to conduct a controlled experiment to quantify the effectiveness of commercially-available and economically affordable UV-C lamps on the reduction of algal concentration on a dominant algal species commonly found in local/regional lakes and rivers. Chlorella vulgaris – a single-cell green algae that is one of the common species in regional freshwater system (Ponnuswamy et al., 2013) – was used in this study. This species can serve as a potential source of food and energy because of its high photosynthetic efficiency (Zelitch, 1971). It can grow at high nitrate and phosphate levels or direct sunlight. By constructing and conducting a controlled experiment, our specific study objectives were to: (1) quantify the magnitude and decreasing trend of algal concentration under different intensities of UV-C radiations; (2) explore if the reduction rate is dependent of contamination level (high vs low concentration) and source of algae (i.e., different variants of the same species); and (3) develop and validate empirical models to predict the changes of algal concentration under different UV-C radiation levels. Results from this study will provide a direct scientific base for designing and engineering a treatment system for selecting the UV-C lamps and treatment duration. We hypothesize that algal concentration under UV-C radiation will decrease exponentially over time, with the reduction rate varying by intensity of UV-C radiation and source of algal population (H1). Additionally, we predict that the reduction rate is dependent of the level of initial algal concentration (i.e., density dependent), with the effectiveness increased for high algal concentration (H2). Logically, UV radiation will cause a higher mortality of algae near the water surface than those in the deeper water because radiation levels decrease with water depth and algal self-shading. This effect with
2. Materials and methods The study was conducted with a full factorial design with three samples of Chlorella vulgaris from two sources. The first algal sample was acquired from the Pilot Anaerobic Digestion/Algal Cultivation Facility (http://www.egr.msu.edu/liao/facilities.htm) of Michigan State University (hereafter named SP1) where the researchers had been growing algae to reduce the nutrients from the water (Chen et al., 2012). They collected algae from a local pond in Okemos, Michigan in 2009 and cultured the algae using anaerobically-digested animal manure for bioenergy production. It reached a steady cultivation rate, the temperature of the water was maintained at approximately 18 °C in order to mimic the temperature of a freshwater pond. Similar to the conditions of the SP2 sample (see text below), the fluorescent lamps provided illumination, except the light was continuous and with more blue radiation. Another sample of the same species was cultivated from a 10-ml test tube of culture ordered online from the Carolina Biological Supply Company (CBSC; http://www.carolina.com/) (hereafter named SP2). Three packs of twelve Alga-Gro concentrated medium tubes (20 ml) from the same company were also ordered to cultivate the algae to grow. A large quantity of tap water was used, but bubbled by air pumps for at least 24 h to eliminate chlorine, which could harm the cultivation process, before adding the medium for culturing the algae (i.e., 20 ml Alga-Gro+1 L H2O per liter of medium per CBSC recommendations). Cultivating the algae required a setup of several growth chambers to account for increase in concentration and volume by starting with a small cup, then gradually increasing the volume to several 6-liter jars, and ending with a standard 10-gallon aquarium fish tank (see CBSC algae cultivation manuals). To create an adequate growing condition, several factors needed to be controlled during the cultivation, including temperature (15–25 °C), pH level, and a lighting schedule. Two sets of standard fluorescent lights (total=4, 40 W each) were used as the light source. The fluorescent lights were powered on for 16 h per day as recommended by the CBSC, and the pH was checked daily and maintained at approximately 7.8 by adding white vinegar to increase the acidity when the water turned to too basic (i.e., a pattern during algal growth). Our cultural process lasted from January 7, 2016 to January 19, 2016. Multiple air pumps were used to continuously mix the water and its dissolved gases (e.g., CO2). Initial UV-C trials indicated that the SP1 culture was more resistant to UV-C radiation than the SP2 culture. The SP1 culture was also denser than the SP2 culture. Therefore, our third source of algal water was an equal mixture of SP1 and SP2. The purpose is to see if there exists significant interactions between the two populations of a specie in responding to the UV-C radiation. With both samples of C. vulgaris cultivated to the desired volume for our experiment ( > 8 gallons), we started our UV-C radiation treatment following a two-way factorial design: 3 levels of UV-C radiation (0 W, 15 W, and 30 W) and three sources of water samples (SP1, SP2, and SP3). Nine 15 W UV-C radiation bulbs (CFL15/UV/MED) were ordered online from the 184
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
Fig. 1. Changes in the mean and standard deviation (SD) algal concentration (%) measured by the Chl RFU (PC s ml−1) based on a 32-h UV-C radiation treatment of three sources (SP1, SP2, and SP3) of algal samples (Chlorella vulgaris) through four replicated treatments (n=4). See Table 1 for the statistical results from analysis of variances (ANOVA).
interpreting the changes in algal concentration. These models were used for each treatment, as well as by sample source and UV-C intensity. Our preliminary modeling results indicated that all three models were reliable with high returns of the correlation coefficient of determination (R2). Given that our long term goal is to find an immediate decrease after UV-C radiation is implemented, we only include the statistical results from the logarithm model in this paper, where value represents the reduction at 1 h after the treatment (i.e., Y=β when X=1). To validate the empirical model on the effects of UV-C radiation on algae concentration, an independent experiment was conducted using the SP1 samples. The SP1 sample was diluted using bubbled tap water and split into three different ratios of water to algae (1:2, 1:1, 2:1). The treatment was made for each sample with 0 W, 15 W, and 30 W for 20 h, with measurements taken every 2 h. Analysis of variance (ANOVA) was applied in several ways to test our hypotheses. First, a 3-way ANOVA was constructed to quantify the significance of water source, UV-C radiation level, and time. Because time is a repeated measure, a repeated two-way ANOVA was applied to quantify the differences by water source (3 sources) and UV-C radiation treatment (3 levels). The results from these two types of ANOVA indicate not only the significance of time (i.e., reduction over time), but also the quantitative contributions of water source and UV-C radiation (H1). To test H2, we quantitatively compared if the changing rate of reduction in ChL with initial density remains the same between the two UV-C radiations, as well as among the three sources of water (i.e., if the slopes are parcel to each other). These statistical analyses were performed using R platform (R Development Core Team 2014, Version 3.1.1).
Norman Lamps Inc. through Amazon.com. Nine 6×12 in. aluminum foil pans were used to hold 1000 ml of algal water samples. Each container was covered with another foil pan (i.e., the lid) to prevent UV-C radiation from escaping the containers. The pans were randomly arranged in 3 rows and 3 columns. The lids were secured to the pans using binder clips to further minimize the UV-C radiation escaping. Holes were created for each container (~1.5 in. in diameter) to insert the UV-C lamps from one side (15 W) or both sides (30 W). The UV-C bulbs were then attached to a wooden frame so that they did not touch the water. The containers were labeled according to their algal water type (SP2, SP1, and SP3) and UV-C radiation intensity (0 W, 15 W, 30 W). Each treatment lasted for 32 h. A YSI 6600 V2 Sonde (https://www.ysi.com/6600-V2-4) was used to measure water properties at 1, 2, 4, 8, 16, and 32 h after UV-C treatments were applied. The primary reason for choosing this widelyused sensor system in water quality studies is because it can provide rapid in situ measurement of many properties of water, including chlorophyll fluorescence, a proxy measurement of algal concentration (Lorenzen, 1966). A disadvantage is that the system does not measure the actual algal cell density in the water, though it can be estimated when the chlorophyll relevant fluorescence unit (Chl RFU) and the actual algal concentration data are available. The sample processing time of three samples takes about 24 min, resulting an approximate 35 h to complete each cycle of the treatments. The same treatments were repeated four times (i.e., n=4). The UV-C radiation was turned off during measurements. A total of eleven variables were recorded with the YSI 6600 Sonde: temperature (°C), species count, water depth, pH, pH mv, turbidity (NTU), ChL (%), Chl RFU, ODO Sat, ODO (mg/l), and BGA-PC (cell/ml) (https://www.ysi.com/6600-V2-4). We used the Chl RFU as a proxy measure for algal concentration. The water temperature with UV-C radiation is about 5–6 °C higher than the reference, causing some loss of water from the container through evaporation. The amount of water left after a complete treatment (32 h) was measured for each container. The evaporative losses were used to recalibrate the algal concentration by assuming a constant evaporative loss for each level of UV-C radiation during the treatment period. Because the initial concentration for each treatment was not the same, we calculated the changes in concentration relative to the initial concentration by dividing the Chl RFU at the time of sampling by the initial condition for each treatment. These values reflect the percentage of algae remaining. These values were used to calculate the mean, standard deviation, minimum and maximum by sample source and UV-C radiation levels. Three existing models within Microsoft Excel were used to construct the empirical relationships between the reduction and treatment duration: Y=α*e-β*X, Y=α*Xβ, and Y=α*ln(X)+β. The coefficients involved in these models have intrinsic meanings in
3. Results UV-C radiation effectively reduced the algal concentration, with clear differences observed by UV-C radiation intensity and source of algal water (Figs. 1–3). The treatment differences are statistically significant amongst the three sources of algal water (p < 0.001) (Table 1); the effect of UV-C intensity also produced significant differences (p < 0.001). Surprisingly, interactions between SP and UV-C intensity were not significant (p=0.314). This suggests that the pattern of reduction is independent of treatment level and source of algal water. When 15 W UV-C radiation was applied to the samples, the mean (STD) algal concentration of SP1, SP2, and SP3 samples within an hour was reduced to 75.3(4.6)%, 51.5(39.0)%, and 70.0(1.0)% of the initial levels, respectively. The concentration was reduced to 69.5(7.6)%, 13.1(1.7)%, and 61.9(3.3)%, respectively, when 30 W UV-C radiation was used during each of the 32-h treatments on the 185
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
Fig. 3. Changes in reductions of algal concentration in the first hour of UV-C radiation treatments (15 W and 30 W) with different initial algal concentrations for: (a) UV radiation intensity, and (b) source of samples. Table 1 Statistical results of repeated measures ANOVA with source of algal-contaminated water (three levels: SP1, SP2, and SP3), UV–C radiation intensity (three levels: 0 W, 15 W, and 30 W), and their interactions. Time (0–32 h) is considered as the repeated measures in our full factorial design with four replications (n=4).
Fig. 2. Changes in algal concentration relative to its initial condition with treatment time under two UV radiation intensities for SP1, SP2, Changes in algal concentration relative to its initial condition with treatment time under two UV radiation intensities (15 W upper line; 30 W lower line) for SP1, SP2, and SP3 samples.and SP3 samples.
three samples. In 8 h, algal concentration of SP2 samples with 15 W and 30 W UV-C radiation decreased to a negligible level of 7.7(10.7)% and 2.5(1.1)%, respectively. These similar reductions (i.e., < 10%), however, were reached in 32 h for the SP3 samples (7.8–8.3%), and never reached for the SP1 samples ( > 18.8%). Overall, the magnitude of the reduction was greater with 30 W UV-C radiation than that with the 15 W UV-C radiation for all three samples, but the magnitude of the reduction was not doubled. The variation among the four replicated treatments seemed the highest for SP1 samples, and the lowest for the SP3 samples (Fig. 1). However, the highest variation was detected for SP2 sample after one hour of the treatment with 15 W UV-C radiation. Interestingly, the algal concentration of the references (i.e., 0 W treatment) showed insignificant changes over the 32-h period, with a slight increase for the SP1 and SP3 samples, but not for the SP2 samples. This suggests possible chlorophyll production for the SP1 and SP3 samples in the dark, but algal mortality for SP2 samples. Algal concentration with UV-C radiation treatment decreased exponentially over time for all three samples, with distinct decreasing trends among the UV-C radiation intensities and the samples (Fig. 2). We modeled each treatment with the exponential, power, and logarithm models embedded in Microsoft Excel and found extremely high predictive powers of all models, with R2 generally greater than 84% (data not shown). The power model produced the highest R2 values, while the exponential model produced the lowest. Applying the logarithm model, the R2 value for the individual treatment (i.e., four
Variable
df
Sum Sq.
Mean Sq.
F-value
P-value
SP UV-C Intensity SP*UV-C Intensity Residuals
2 2 4 243
21,850 202,315 3248 165,401
10,925 101,157 812 681
16.051 148.616 1.193
< 0.001 < 0.001 0.314
replications) ranged 96.7–99.2%, 73.1–98.3%, and 96.2–99.3% for SP1, SP2, and SP3 samples, respectively, when 15 W UV-C radiations were applied. The R2 was 95.4–99.4%, 69.8–75.2%, and 96.0–99.3%, respectively, when 30 W UV-C radiations were used. When all results from the four replications were pooled, the predictive power for SP1 sample was 89.0% and 92.7% for 15 W and 30 W UV-C radiation intensity, respectively (Fig. 2a), 73.2% and 72.3% for SP2 samples (Fig. 2b), and 96.0% and 95.8% for the SP3 samples (Fig. 2c). Our hypothesis that the effectiveness of UV-C treatment on algal concentration is dependent of initial concentration was also accepted (Fig. 3). The reduction in the first hour (i.e., R2 value in the logarithm model) of the UV-C radiation treatment appeared positively correlated with the initial algal concentration (Fig. 3a) for both UV-C radiation intensities. A greater density dependency was apparent for the 30 W UV-C treatment (i.e., steeper slope) than that with 15 W UV-C intensity (Fig. 3a). Interestingly, this density-dependency was not detected among the three samples, although the reduction in the first hour was the highest for SP1 samples and the lowest for the SP3 samples (Fig. 3b). The modeled reduction in algal concentration based on the repeated treatments matched the results of an independent experiment very well, with an overall R2 of 97.4% for 15 W UV-C treatment, and 98.6% for the 30 W treatment (Fig. 4). However, our models overestimated the reduction by 19.3% and 22.4% for the two intensities. These 186
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
chemical or biological agents. A NBS-based approach in treating polluted water is needed. Our endeavors in seeking an environmentally-friendly method for treating algal contaminated water for purification stimulated this study and the corresponding experiment. It has been widely recognized that UV-C radiation can cause the mortality of many microorganisms (e.g., bacterial, fungi, and algae) in both air and water (Takeguchi and Oyobe, 1983). However, quantitative description for the reduction of algae with different UV-C radiation intensities has not been reported in the literature. With UV-C treatment, one can expect that the concentration of these organisms in the air and water will be significantly decreased. Here, we hypothesized that the reduction will be exponential with the duration of the treatment (H1). This reduction process, however, may vary by the agent and its amount due to different resistances (H2). Both hypotheses are confirmed with our experimental data (Fig. 2 for H1; and Fig. 3 for H2). All three samples of the algae-contaminated water showed exponential decreases in their concentrations. For SP2 sample, approximately 50% of the algae was reduced with a 15 W UV-C radiation lamp within an hour. With 30 W UV-C lamp, up to 87% algae can be reduced in an hour. For the more UV-resistant SP1 samples, however, this can take more than 32 h regardless of being the same species (Fig. 1). Clearly, the reduction is dependent of the agent (i.e., variants of the same species). To validate the effectiveness and potential application of our treatment system, we constructed a field chamber using four 15 W UV-C bulbs in a 0.6×2 m chamber at the Beijing Botanical Garden in July of 2016. For the treatments, the field chamber was filled with lake water and observations were recorded every 2 h for 12 h. Due to the lack of equipment, only true color photos were taken to see the effects of the UV-C treatment. After 12 h of treatment, it was observed that the water within the treatment system was clear. The observations were consistent with our experimental results (Figs. 1–3) throughout three replications. More importantly, we took one sample prior and one after 1 h of treatment to compare and examine cell structures under a
Fig. 4. Relationships between predicted algal concentration (Chl RFU) (1000×PC s ml−1) and measured values from an independent experiment of UV-C radiation treatments with different initial conditions using the SP1 samples.
overestimations seemed more pronounced when algal concentration was high. Inclusion of initial density into the model, based on the empirical relationships in Fig. 3a will be a necessity to improve the accuracies of the models.
4. Discussion With increasing and more widespread of algal blooms in inland water systems, one of the forefront needs is the clear scientific basis when faced with public panic and institutional pressure. One of the most urgent needs to address this pressing issue is to seek naturebased, effective solutions to reduce algal concentration. In order to maintain the quality of the water, it is safest to not resort to the use of
Fig. 5. Microscopic photos (true color) of natural water with algal blooms in a lake of the Beijing Botanical Garden before and after 1 h treatment with 15 W UV-C radiation. Algae cells appeared to be ruptured after the treatment. These photos were taken at 2 spatial resolutions to demonstrate the breakdown of algal cells.
187
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
of toxins without rupturing the cells (Ou et al., 2011). In regards to effectively applying our results in freshwater lakes, an apparatus can be built and installed onto floatation devices (e.g., boats, buoys). To increase the engineering design of a treatment system, one may also need to consider circulating the water so that UV-C radiation can produce the strongest effects by evenly striking all parts of the water. Following the NBS concept, the treatment system can be solarpower-based. For example, in our trial treatment system that was applied in the Beijing Botanical Garden, a volume of 0.24 m3 of water was successfully cleaned in an hour (Fig. 5). Over a 24-h period, 5.76 m3 of water can be cleaned. If this similar solar-powered system is installed onto various floatation devices over Lake Erie, a large amount of water can be hypothetically cleaned daily (57.6 m3 per day with 10 systems). Because the treatment system is low-cost and solar-powered, this suggests that the deployment of this system over bodies of freshwater – even large lakes such as Lake Erie – is achievable. It is also possible that a large treatment facility can be built, but our objective in this study was to provide a scientific base for designing small-scale, portable treatment systems.
microscope (Fig. 5). It appeared that the cells had been destructed as a result of the UV-C treatment. This indicates that the algae was killed within one hour, despite that the sample remained green. Although the green color had disappeared within 2–4 h of treatment (Figs. 1–3), it is most likely that the cells had already died prior to the discoloration. Although UV-C radiation is an effective method of reducing algal concentration, other variables such as temperature may affect its effectiveness. For example, in a study by Wong et al. (2015), it was discovered that increasing temperatures could counteract the effectiveness of radiation. Different levels of UV-C radiation treatments and temperatures on C. vulgaris were administered in the study and it was concluded that UV-C radiation intensity was independent of changes in temperature. Although UV-C radiation effectively reduced algal concentrations, increased temperatures led to an improved ability for damaged cells to recover from UVR stress. This could potentially pose an issue for UV treatment systems when lakes or rivers rise in temperatures due to the effects of global warming. The underline mechanisms for the differences remain to be explored with consideration of the cultivation history (e.g., the media) within species variation and other potential biological and chemical properties of the water (nutrients, pH level, etc). The fact that a different response from SP3 samples to the same UVC treatment (Fig. 2c) from SP1 and SP2 samples also stimulate another potential hypothesis about the role of species interactions (e.g., elevated competition under stress; Jeffries and Mills, 1990) in responding to the UV radiation. Some natural following up steps would be analyzing the quantitative contributions of interactive casual relationships by algal density, different species of algae from in situ water sources, and UV-C radiation intensity through analysis of variance in more robust experiments (e.g., large replications). Future efforts can be further enhanced by mechanistic exploration of altered species interactions, resource limitations, population dynamics, nutrient cycling, species-environment relationships under different UV-C conditions, etc. One of the concerns for applying our approach on algae-polluted water in the field is that nutrients will be returned to the water, which will continue promoting the growth of algal population. However, our treatment will reduce the algal concentration, temporarily disrupting the growth. When applied properly in time and space, the treatment can mitigate the magnitude and scale of algal blooms (Ouyang et al., 2017). Another uncertainty is the resistance of damaged cells as a result of UV treatment. Algae could potentially build up resistance to UV radiation (Karsten and Holzinger, 2014), which is demonstrated in algae-UV relationships in alpine soils. In addition, algae responses to UV is also in condition to other variables (Wong et al., 2015). This suggests that future efforts are needed to examine the effects of confounding variables under varying environments from physiological and genetic perspectives (Ou et al., 2011; Brennan and Collins, 2015; McGivney et al., 2015; Sharma et al., 2015a, 2015b; Wong et al., 2015). A plausible finding of this study is the highly-reliable models that can be used for predicting the changes in algal concentration in the water under different UV-C radiation intensity and initial algal concentration (Figs. 1–3). The species we used in this study is a common species in many inland lakes (e.g., Lake Erie; Chaffin et al., 2014) and elsewhere (Ponnuswamy et al., 2013), suggesting that the model can be readily applied for treating the water from these lakes. Although these models were constructed for single species and in a controlled environment, we speculate the reduction rate will be at similar level for the natural lake water, where multiple species coexist. With this assumption, one can determine the amount of time for their treatment goals. However, our finding that the effects are densitydependent infers that treatment time needs to be set based on the algal concentration, with stronger UV-C radiation needed for high concentration water (Fig. 3). When studying other algal species in the future that may be toxic, such as Microcystis aeruginosa, it is imperative that we quantify an appropriate UV-C dosage that will degrade the release
5. Conclusions Both hypotheses regarding the effects of different UV-C radiation intensities on different algal concentrations and densities were accepted. Different intensities of UV-C radiation were applied to variations of Chlorella vulgaris. The 30 W UV-C radiation had a stronger effect than the 15 W UV-C radiation on reducing algal concentration. Another finding was that the UV-C radiations had the greatest impact on SP2 samples, and the lowest impact on SP1 samples. Furthermore, an independent study was conducted with only SP1 samples, but at different densities by diluting the samples with water. The results showed that the 30 W UV-C radiation still had the greatest impact and that as density increased, the reduction rates decreased. The independent experiment confirmed the density-dependent factor in the previous study. For each treatment, as well as UV-C radiation intensities and sample sources, empirical models were developed with high confidences (R2 > 84%). Using the information from the research and other similar future references, possible treatment systems could potentially be constructed and implemented indoors as well as in the fields. Acknowledgements We would like to thank many individuals for their advice, support and time for this project. Carol Stepien and Chris Mayer provided the essential equipment, including aquarium fish tanks, sets of fluorescent lights and air pumps to cultivate the algal growth; Wei Liao of the Michigan State University supplied a large sample of algae from his lab; Dr. Shenghua Gao and Xudong Zhang helped to construct a field pilot treatment chamber to validate the in situ application of the method at the Beijing Botanical Garden; Yao Cheng of the Chinese Academy of Sciences assisted in operating the microscope at IBCAS for changes at cell level. Two anonymous reviewers and the handling editor provided constructive suggestions to improved the quality of this paper. References Borderie, F., Alaoui-Sehmer, L., Bousta, F., Alaoui-Sossé, B., 2014. Cellular and molecular damage caused by high UV-C irradiation of the cave-harvested green alga Chlorella minutissima: implications for cave management. Int. Biodeterior. Biodegrad. 93, 118–130. Brennan, G., Collins, S., 2015. Growth responses of a green alga to multiple environmental divers. Nat. Clim. Change. http://dx.doi.org/10.1038/ NCLIMATE2682. Chaffin, J.D., Bridgeman, T.B., Bade, D.L., Mobilian, C.N., 2014. Summer phytoplankton nutrient limitation in Maumee Bay of Lake Erie during high-flow and low-flow years. J. Gt. Lakes Res. 40 (3), 524–531. Chen, R., Li, R., Dietz, L., Liu, Y., Stevenson, J., Liao, W., 2012. Freshwater algal
188
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
Conserv. Lett.. http://dx.doi.org/10.1111/conl.12216. McGivney, E., Carlsson, M., Gustafsson, J.P., Gorokhova, E., 2015. Effects of UV-C and vacuum-UV TiO2 advanced oxidation processes on the acute mortality of microalgae. Photochem. Photobiol. 91, 1142–1149. Michalak, A.M., Anderson, E.J., Beletsky, D., Boland, S., Bosch, N.S., Bridgeman, T.B., DePinto, J.V., 2013. Record-setting algal bloom in Lake Erie caused by agricultural and meteorological trends consistent with expected future conditions. Proc. Natl. Acad. Sci. USA 110 (16), 6448–6452. Ou, H., Gao, N., Deng, Y., Qiao, J., Wang, H., 2011. Immediate and long-term impacts of UV-C irradiation on photosynthetic capacity, survival and microcystin-LR release risk of Microcystis aeruginosa. Water Res. 46 (2012), 1241–1250. Ouyang, Z., Shao, C., Chu, H., Becker, R., Bridgeman, T., Stepien, C.A., John, R., Chen, J., 2017. Algal bloom reduced carbon emission to the atmosphere in western Lake Erie. Remote Sens. 9 (1), 44. http://dx.doi.org/10.3390/rs9010044. Pogge, J., 2015. Anthropogenic Drivers of Harmful Algal Blooms, Specifically Microcystin, from Fresh to Marine Environments. Capstone, Nova Southeast University. Ponnuswamy, I., Madhavan, S., Shabudeen, S., 2013. Isolation and characterization of green microalgae for carbon sequestration, waste water treatment and bio-fuel production. Int. J. Bio-Sci. Bio-Technol. 5, 17–26. Sharma, K.K., Li, Y., Schenk, P.M., 2015b. Rapid lipid induction in Chlorella sp. by UV-C radiation. Bioenergy Res. 8, 1824–1830. Sharma, K.K., Ahmed, F., Schenk, P.M., Li, Y., 2015a. UV-C mediated rapid carotenoid induction and settling performance of Dunaliella salina and Haematococcus pluvialis. Biotechnol. Bioeng. 112, 2106–2114. Takeguchi, M., Oyobe, I., 1983. U.S. Patent No. 4,390,432. Washington, DC, U.S. Patent and Trademark Office. Wong, C.-Y., Teoh, M.-L., Phang, S.-M., Lim, P.-E., 2015. Interactive effects of temperature and UV radiation on photosynthesis of Chlorella strains from polar, temperate, and tropical environments: differential impacts on damage and repair. PLoS One 10 (10), e0139469. http://dx.doi.org/10.1371/journal.pone.0139469. Zelitch, I., 1971. Photosynthesis, Photorespiration and Plant Productivity. Academic Press, London, UK, 275.
cultivation with animal waste for nutrient removal and biomass accumulation. J. Biomass Bioenergy 39, 128–138. Coppen, S., 2015. Drivers of Harmful Algal Blooms in the North Sea (M.S. thesis). University of Gent. Hall, K.K., Giannetta, E.T., Getchell-White, S.I., Durbin, N.J., Farr, B.M., 2003. Ultraviolet light disinfection of hospital water for preventing nosocomial Legionella infection: a 13-year follow-up. Infect. Control Hosp. Epidemiol. 24 (8), 580–583. Hallegraeff, G.M., 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32 (2), 79–99. Heisler, J., Glibert, P.M., Burkholder, J.M., Anderson, D.M., Cochlan, W., Dennison, W.C., Dortch, Q., Gobler, C.J., Heil, 1, C.A., Humphries, E., Lewitus, A., Magnien, R., Marshall, H.G., Sellner, K., Stockwell, D.A., Stoecker, D.K., Suddleson, M., 2008. Eutrophication and harmful algal blooms: a scientific consensus. Harmful Algae 8 (1), 1–13. Ho, J.C., Michalak, A.M., 2015. Challenges in tracking harmful algal blooms: a synthesis of evidence from Lake Erie. J. Gt. Lakes Res. 41, 317–325. Jeffries, M., Mills, D., 1990. Freshwater Ecology: Principles and Applications. Belhaven Press, London. Jetoo, S., Grover, V.I., Krantzerg, G., 2015. The Toledo drinking water advisory: suggested application of the water safety planning approach. Sustainability 7 (8), 9787–9808. Kaplan, A., Badger, M.R., Berry, J.A., 1980. Photosynthesis and the intracellular inorganic carbon pool in the bluegreen alga Anabaena variabilis: response to external CO2 concentration. Planta 149 (3), 219–226. Karsten, U., Holzinger, A., 2014. Green algae in alpine biological soil crust communities: acclimation strategies against ultraviolet radiation and dehydration. Biodivers. Conserv. 23, 1845–1858. Loftin, K.A., Clark, J.M., Journey, C.A., Kolpin, D.W., Van Metre, P.C., Bradley, P.M., 2016. Spatial and temporal variation in microcystins occurrence in wadeable streams in the southeastern USA. Environ. Toxicol. Chem. 35. http://dx.doi.org/10.1002/ etc.3391. Lorenzen, C.J., 1966. A method for the continuous monitoring of in-vivo chlorophyll concentration. Deep Sea Res. 13, 223–227. Maes, J., Jacobs, S., 2015. Nature-based solutions for Europe's sustainable development.
189
Contents lists available at ScienceDirect
Environmental Research journal homepage: www.elsevier.com/locate/envres
The reduction of Chlorella vulgaris concentrations through UV-C radiation treatments: A nature-based solution (NBS) Erika S. Chena, Thomas B. Bridgemanb, a b
MARK
⁎
Sylvania Northview High School, Sylvania, OH 43560, USA Lake Erie Center, University of Toledo, Oregon, OH 43616, USA
A R T I C L E I N F O
A BS T RAC T
Keywords: Algal concentration UV radiation HAB Mortality Density-dependent NBS Modeling
Algal blooms have become a pressing issue in inland freshwater systems on local and global scales. A plausible approach to reducing algae without the use of chemical/biological agents is through the use of UV-C radiation from lamps potentially powered by in situ solar panels to eliminate algae. Yet, the quantitative scientific base has not been established. Our objective is to conduct a controlled experiment to quantify the effectiveness of UV-C radiation on the reduction of Chlorella vulgaris, a common algal species in the Great Lakes region. A full factorial design of three intensities of UV-C radiation (0, 15, and 30 W) and three sources of C. vulgaris was constructed to test the corresponding hypotheses. Empirical models were constructed to predict the reductions. UV-C radiation effectively reduced the algal concentration with clear differences by radiation level and source of algal water. Algal concentration decreased exponentially over time, with distinct decreasing trends among the radiation intensities and the samples. With 15 W UV-C radiation, algal concentration of three samples were reduced to 75.3%, 51.5%, and 70.0% of the initial level within an hour, respectively. We also found a clear density-dependent reduction rate by UV radiation. Using this information, more efficient treatment systems could be constructed and implemented for cleaning algae-contaminated water.
1. Introduction Algal blooms have become a major environmental issue in many inland lakes (Pogge, 2015) and are major concerns of the general public at a regional, national, and global level (Hallegraeff, 1993; Ho and Michalak, 2015). In the Midwest region of the United States, Lake Erie has had a history of harmful algal blooms (HABs), with a maximum level in the summer of 2015 (cf Ouyang et al., 2017). Another algal bloom event of a very similar magnitude in Lake Erie erupted in 2011 – the largest algal bloom in the lake's history up to that point, which had a peak over three times greater than all previous algal blooms. Like the 2015 algal bloom, the main cause was eutrophication. Intensified agricultural practices and meteorological conditions in the spring were directly related with increased phosphorus depositions in the western basin of the lake (Chaffin et al., 2014). Other factors contributing to the issue were the extended period of weak lake circulation, which prevented the flushing of nutrients from the system (Michalak et al., 2013). Although not as large as the HABs of 2011 and 2015, the 2014 algal bloom still had a great impact on the general public. More than 400,000 people were without safe drinking water for 2.5 days during the 2014 blooms (Jetoo et al., 2015).
⁎
Corresponding author. E-mail address: [email protected] (T.B. Bridgeman).
http://dx.doi.org/10.1016/j.envres.2017.03.007 Received 2 January 2017; Received in revised form 2 March 2017; Accepted 4 March 2017 0013-9351/ © 2017 Elsevier Inc. All rights reserved.
HABs are not restricted to the Great Lakes. A similar case was found in the Southeastern United States. From 1993 to 2000, there were observations of potential microcystin-producing cyanobacteria in 74% of streams in Alabama, Georgia, South Carolina, and North Carolina (Loftin et al., 2016). In the North Sea, HABs have had many negative impacts, including poisoning, death, economic damage to aquaculture, and decline in tourism. The area has observed a strong increase in algal blooms since the Industrial Revolution. The main causes of the algal blooms were ballast water dispersal, eutrophication, and global warming. However, nutrients were not the biggest factor in determining the occurrence of HABs. Instead, the main factors were the initial density and interspecific competition (Coppen, 2015). Another key factor affecting algae growth is the supply of carbon dioxide (CO2). In a study quantifying the relationship of the photosynthesis of algae (Anabaena variabilis) to CO2 concentration, the author indicated that as the CO2 concentration increased, the photosynthetic properties of algae were greater (Kaplan et al., 1980). Consequently, developing environmentally safe and economically affordable methods and systems has become a challenge in cleaning up contaminated freshwater (Heisler et al., 2008). Recently developed concepts and principles of nature-based solutions (NBS) (Maes and
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
water depth will be elevated as the surface water gets more transparent due to reduction of algae concentration near the surface. Consequently, the total algal mortality in the entire water column will not be maintained over time, but will decrease in effectiveness, resulting in exponential changes in mortality (i.e., H1). Such a change is also dependent of the intensity of UV-C radiation, with a higher radiation causing more mortality. The effectiveness of UV-C radiation may also vary by the source of algae if the algae were cultivated using different lights and media (i.e., their resistances to UV-C radiation are not the same). Finally, one would expect that the same level of UV-C radiation will cause more algae to die when the algal concentration is high (H2). Consequently, the relative reduction in algal concentration would be higher for water with high algal concentration than that with low concentration (H2).
Jacobs, 2015) appear plausible and stimulating for the scientific community in seeking a NBS for treating polluted water. Based on this concept, we sought an approach of applying UV that can be created in the field using solar panels on algae-polluted water for reducing algal abundance. Our long-term study objective is to develop a scientifically-sound, portable and user-friendly system that can be used to treat algaecontaminated water. A NBS approach to the matter is through the use of UV radiation. Germicidal UV (UV-C) radiation has been proven to possess the ability to eliminate algae and other microorganisms. UV-C radiation is able to eliminate such microorganisms by deactivating the DNA and damaging the nucleic acids. UV-C radiation has not only been effective in treatment, but is also inexpensive, low maintenance, and simple. In 1983, a design for a water purifying tank was created that incorporated UV-C radiation and biological oxidation. The UV-C radiation was able to inhibit the growth of bacteria (Takeguchi and Oyobe, 1983). UV-C radiation has also been used in disinfecting hospitals. In a 13-year study, a UV apparatus was placed in the water main of a new hospital building. Before the construction of the new hospital, 27% of water samples taken from the old hospital contained Legionella. After 13 years, the UV-C radiation usage resulted in negative water cultures and no cases of nosocomial Legionella pneumophila (Hall et al., 2003). This study demonstrated the effectiveness of UV-C radiation. Along with microorganisms such as bacteria and mold, UV-C radiation has also been proven to eliminate harmful algal species. For example, when exposed to UV-C treatments for 21 days, treatments of Chlorella minutissima experienced chlorophyll bleaching, oxidative stress, programmed cell death, and inhibited metabolic activity (Borderie et al., 2014). UV-C is also capable of reducing microcystin-LR (MC-LR) release risk of Microcystis aeruginosa, the primary cause of Lake Erie HABs (Ou et al., 2011). As UV-C dosages increased, the photosynthetic capacities of M. aeruginosa were reduced, accompanied by a complete degradation of MC-LR immediately after irradiation. Unfortunately, quantifying the applications of UV-C radiation to Chlorella vulgaris has not been studied extensively. Our literature search did not yield any publications that have the scientific base for determining the treatment duration for common algal species in our freshwater lakes and rivers in the Great Lakes region. Our objective is to conduct a controlled experiment to quantify the effectiveness of commercially-available and economically affordable UV-C lamps on the reduction of algal concentration on a dominant algal species commonly found in local/regional lakes and rivers. Chlorella vulgaris – a single-cell green algae that is one of the common species in regional freshwater system (Ponnuswamy et al., 2013) – was used in this study. This species can serve as a potential source of food and energy because of its high photosynthetic efficiency (Zelitch, 1971). It can grow at high nitrate and phosphate levels or direct sunlight. By constructing and conducting a controlled experiment, our specific study objectives were to: (1) quantify the magnitude and decreasing trend of algal concentration under different intensities of UV-C radiations; (2) explore if the reduction rate is dependent of contamination level (high vs low concentration) and source of algae (i.e., different variants of the same species); and (3) develop and validate empirical models to predict the changes of algal concentration under different UV-C radiation levels. Results from this study will provide a direct scientific base for designing and engineering a treatment system for selecting the UV-C lamps and treatment duration. We hypothesize that algal concentration under UV-C radiation will decrease exponentially over time, with the reduction rate varying by intensity of UV-C radiation and source of algal population (H1). Additionally, we predict that the reduction rate is dependent of the level of initial algal concentration (i.e., density dependent), with the effectiveness increased for high algal concentration (H2). Logically, UV radiation will cause a higher mortality of algae near the water surface than those in the deeper water because radiation levels decrease with water depth and algal self-shading. This effect with
2. Materials and methods The study was conducted with a full factorial design with three samples of Chlorella vulgaris from two sources. The first algal sample was acquired from the Pilot Anaerobic Digestion/Algal Cultivation Facility (http://www.egr.msu.edu/liao/facilities.htm) of Michigan State University (hereafter named SP1) where the researchers had been growing algae to reduce the nutrients from the water (Chen et al., 2012). They collected algae from a local pond in Okemos, Michigan in 2009 and cultured the algae using anaerobically-digested animal manure for bioenergy production. It reached a steady cultivation rate, the temperature of the water was maintained at approximately 18 °C in order to mimic the temperature of a freshwater pond. Similar to the conditions of the SP2 sample (see text below), the fluorescent lamps provided illumination, except the light was continuous and with more blue radiation. Another sample of the same species was cultivated from a 10-ml test tube of culture ordered online from the Carolina Biological Supply Company (CBSC; http://www.carolina.com/) (hereafter named SP2). Three packs of twelve Alga-Gro concentrated medium tubes (20 ml) from the same company were also ordered to cultivate the algae to grow. A large quantity of tap water was used, but bubbled by air pumps for at least 24 h to eliminate chlorine, which could harm the cultivation process, before adding the medium for culturing the algae (i.e., 20 ml Alga-Gro+1 L H2O per liter of medium per CBSC recommendations). Cultivating the algae required a setup of several growth chambers to account for increase in concentration and volume by starting with a small cup, then gradually increasing the volume to several 6-liter jars, and ending with a standard 10-gallon aquarium fish tank (see CBSC algae cultivation manuals). To create an adequate growing condition, several factors needed to be controlled during the cultivation, including temperature (15–25 °C), pH level, and a lighting schedule. Two sets of standard fluorescent lights (total=4, 40 W each) were used as the light source. The fluorescent lights were powered on for 16 h per day as recommended by the CBSC, and the pH was checked daily and maintained at approximately 7.8 by adding white vinegar to increase the acidity when the water turned to too basic (i.e., a pattern during algal growth). Our cultural process lasted from January 7, 2016 to January 19, 2016. Multiple air pumps were used to continuously mix the water and its dissolved gases (e.g., CO2). Initial UV-C trials indicated that the SP1 culture was more resistant to UV-C radiation than the SP2 culture. The SP1 culture was also denser than the SP2 culture. Therefore, our third source of algal water was an equal mixture of SP1 and SP2. The purpose is to see if there exists significant interactions between the two populations of a specie in responding to the UV-C radiation. With both samples of C. vulgaris cultivated to the desired volume for our experiment ( > 8 gallons), we started our UV-C radiation treatment following a two-way factorial design: 3 levels of UV-C radiation (0 W, 15 W, and 30 W) and three sources of water samples (SP1, SP2, and SP3). Nine 15 W UV-C radiation bulbs (CFL15/UV/MED) were ordered online from the 184
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
Fig. 1. Changes in the mean and standard deviation (SD) algal concentration (%) measured by the Chl RFU (PC s ml−1) based on a 32-h UV-C radiation treatment of three sources (SP1, SP2, and SP3) of algal samples (Chlorella vulgaris) through four replicated treatments (n=4). See Table 1 for the statistical results from analysis of variances (ANOVA).
interpreting the changes in algal concentration. These models were used for each treatment, as well as by sample source and UV-C intensity. Our preliminary modeling results indicated that all three models were reliable with high returns of the correlation coefficient of determination (R2). Given that our long term goal is to find an immediate decrease after UV-C radiation is implemented, we only include the statistical results from the logarithm model in this paper, where value represents the reduction at 1 h after the treatment (i.e., Y=β when X=1). To validate the empirical model on the effects of UV-C radiation on algae concentration, an independent experiment was conducted using the SP1 samples. The SP1 sample was diluted using bubbled tap water and split into three different ratios of water to algae (1:2, 1:1, 2:1). The treatment was made for each sample with 0 W, 15 W, and 30 W for 20 h, with measurements taken every 2 h. Analysis of variance (ANOVA) was applied in several ways to test our hypotheses. First, a 3-way ANOVA was constructed to quantify the significance of water source, UV-C radiation level, and time. Because time is a repeated measure, a repeated two-way ANOVA was applied to quantify the differences by water source (3 sources) and UV-C radiation treatment (3 levels). The results from these two types of ANOVA indicate not only the significance of time (i.e., reduction over time), but also the quantitative contributions of water source and UV-C radiation (H1). To test H2, we quantitatively compared if the changing rate of reduction in ChL with initial density remains the same between the two UV-C radiations, as well as among the three sources of water (i.e., if the slopes are parcel to each other). These statistical analyses were performed using R platform (R Development Core Team 2014, Version 3.1.1).
Norman Lamps Inc. through Amazon.com. Nine 6×12 in. aluminum foil pans were used to hold 1000 ml of algal water samples. Each container was covered with another foil pan (i.e., the lid) to prevent UV-C radiation from escaping the containers. The pans were randomly arranged in 3 rows and 3 columns. The lids were secured to the pans using binder clips to further minimize the UV-C radiation escaping. Holes were created for each container (~1.5 in. in diameter) to insert the UV-C lamps from one side (15 W) or both sides (30 W). The UV-C bulbs were then attached to a wooden frame so that they did not touch the water. The containers were labeled according to their algal water type (SP2, SP1, and SP3) and UV-C radiation intensity (0 W, 15 W, 30 W). Each treatment lasted for 32 h. A YSI 6600 V2 Sonde (https://www.ysi.com/6600-V2-4) was used to measure water properties at 1, 2, 4, 8, 16, and 32 h after UV-C treatments were applied. The primary reason for choosing this widelyused sensor system in water quality studies is because it can provide rapid in situ measurement of many properties of water, including chlorophyll fluorescence, a proxy measurement of algal concentration (Lorenzen, 1966). A disadvantage is that the system does not measure the actual algal cell density in the water, though it can be estimated when the chlorophyll relevant fluorescence unit (Chl RFU) and the actual algal concentration data are available. The sample processing time of three samples takes about 24 min, resulting an approximate 35 h to complete each cycle of the treatments. The same treatments were repeated four times (i.e., n=4). The UV-C radiation was turned off during measurements. A total of eleven variables were recorded with the YSI 6600 Sonde: temperature (°C), species count, water depth, pH, pH mv, turbidity (NTU), ChL (%), Chl RFU, ODO Sat, ODO (mg/l), and BGA-PC (cell/ml) (https://www.ysi.com/6600-V2-4). We used the Chl RFU as a proxy measure for algal concentration. The water temperature with UV-C radiation is about 5–6 °C higher than the reference, causing some loss of water from the container through evaporation. The amount of water left after a complete treatment (32 h) was measured for each container. The evaporative losses were used to recalibrate the algal concentration by assuming a constant evaporative loss for each level of UV-C radiation during the treatment period. Because the initial concentration for each treatment was not the same, we calculated the changes in concentration relative to the initial concentration by dividing the Chl RFU at the time of sampling by the initial condition for each treatment. These values reflect the percentage of algae remaining. These values were used to calculate the mean, standard deviation, minimum and maximum by sample source and UV-C radiation levels. Three existing models within Microsoft Excel were used to construct the empirical relationships between the reduction and treatment duration: Y=α*e-β*X, Y=α*Xβ, and Y=α*ln(X)+β. The coefficients involved in these models have intrinsic meanings in
3. Results UV-C radiation effectively reduced the algal concentration, with clear differences observed by UV-C radiation intensity and source of algal water (Figs. 1–3). The treatment differences are statistically significant amongst the three sources of algal water (p < 0.001) (Table 1); the effect of UV-C intensity also produced significant differences (p < 0.001). Surprisingly, interactions between SP and UV-C intensity were not significant (p=0.314). This suggests that the pattern of reduction is independent of treatment level and source of algal water. When 15 W UV-C radiation was applied to the samples, the mean (STD) algal concentration of SP1, SP2, and SP3 samples within an hour was reduced to 75.3(4.6)%, 51.5(39.0)%, and 70.0(1.0)% of the initial levels, respectively. The concentration was reduced to 69.5(7.6)%, 13.1(1.7)%, and 61.9(3.3)%, respectively, when 30 W UV-C radiation was used during each of the 32-h treatments on the 185
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
Fig. 3. Changes in reductions of algal concentration in the first hour of UV-C radiation treatments (15 W and 30 W) with different initial algal concentrations for: (a) UV radiation intensity, and (b) source of samples. Table 1 Statistical results of repeated measures ANOVA with source of algal-contaminated water (three levels: SP1, SP2, and SP3), UV–C radiation intensity (three levels: 0 W, 15 W, and 30 W), and their interactions. Time (0–32 h) is considered as the repeated measures in our full factorial design with four replications (n=4).
Fig. 2. Changes in algal concentration relative to its initial condition with treatment time under two UV radiation intensities for SP1, SP2, Changes in algal concentration relative to its initial condition with treatment time under two UV radiation intensities (15 W upper line; 30 W lower line) for SP1, SP2, and SP3 samples.and SP3 samples.
three samples. In 8 h, algal concentration of SP2 samples with 15 W and 30 W UV-C radiation decreased to a negligible level of 7.7(10.7)% and 2.5(1.1)%, respectively. These similar reductions (i.e., < 10%), however, were reached in 32 h for the SP3 samples (7.8–8.3%), and never reached for the SP1 samples ( > 18.8%). Overall, the magnitude of the reduction was greater with 30 W UV-C radiation than that with the 15 W UV-C radiation for all three samples, but the magnitude of the reduction was not doubled. The variation among the four replicated treatments seemed the highest for SP1 samples, and the lowest for the SP3 samples (Fig. 1). However, the highest variation was detected for SP2 sample after one hour of the treatment with 15 W UV-C radiation. Interestingly, the algal concentration of the references (i.e., 0 W treatment) showed insignificant changes over the 32-h period, with a slight increase for the SP1 and SP3 samples, but not for the SP2 samples. This suggests possible chlorophyll production for the SP1 and SP3 samples in the dark, but algal mortality for SP2 samples. Algal concentration with UV-C radiation treatment decreased exponentially over time for all three samples, with distinct decreasing trends among the UV-C radiation intensities and the samples (Fig. 2). We modeled each treatment with the exponential, power, and logarithm models embedded in Microsoft Excel and found extremely high predictive powers of all models, with R2 generally greater than 84% (data not shown). The power model produced the highest R2 values, while the exponential model produced the lowest. Applying the logarithm model, the R2 value for the individual treatment (i.e., four
Variable
df
Sum Sq.
Mean Sq.
F-value
P-value
SP UV-C Intensity SP*UV-C Intensity Residuals
2 2 4 243
21,850 202,315 3248 165,401
10,925 101,157 812 681
16.051 148.616 1.193
< 0.001 < 0.001 0.314
replications) ranged 96.7–99.2%, 73.1–98.3%, and 96.2–99.3% for SP1, SP2, and SP3 samples, respectively, when 15 W UV-C radiations were applied. The R2 was 95.4–99.4%, 69.8–75.2%, and 96.0–99.3%, respectively, when 30 W UV-C radiations were used. When all results from the four replications were pooled, the predictive power for SP1 sample was 89.0% and 92.7% for 15 W and 30 W UV-C radiation intensity, respectively (Fig. 2a), 73.2% and 72.3% for SP2 samples (Fig. 2b), and 96.0% and 95.8% for the SP3 samples (Fig. 2c). Our hypothesis that the effectiveness of UV-C treatment on algal concentration is dependent of initial concentration was also accepted (Fig. 3). The reduction in the first hour (i.e., R2 value in the logarithm model) of the UV-C radiation treatment appeared positively correlated with the initial algal concentration (Fig. 3a) for both UV-C radiation intensities. A greater density dependency was apparent for the 30 W UV-C treatment (i.e., steeper slope) than that with 15 W UV-C intensity (Fig. 3a). Interestingly, this density-dependency was not detected among the three samples, although the reduction in the first hour was the highest for SP1 samples and the lowest for the SP3 samples (Fig. 3b). The modeled reduction in algal concentration based on the repeated treatments matched the results of an independent experiment very well, with an overall R2 of 97.4% for 15 W UV-C treatment, and 98.6% for the 30 W treatment (Fig. 4). However, our models overestimated the reduction by 19.3% and 22.4% for the two intensities. These 186
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
chemical or biological agents. A NBS-based approach in treating polluted water is needed. Our endeavors in seeking an environmentally-friendly method for treating algal contaminated water for purification stimulated this study and the corresponding experiment. It has been widely recognized that UV-C radiation can cause the mortality of many microorganisms (e.g., bacterial, fungi, and algae) in both air and water (Takeguchi and Oyobe, 1983). However, quantitative description for the reduction of algae with different UV-C radiation intensities has not been reported in the literature. With UV-C treatment, one can expect that the concentration of these organisms in the air and water will be significantly decreased. Here, we hypothesized that the reduction will be exponential with the duration of the treatment (H1). This reduction process, however, may vary by the agent and its amount due to different resistances (H2). Both hypotheses are confirmed with our experimental data (Fig. 2 for H1; and Fig. 3 for H2). All three samples of the algae-contaminated water showed exponential decreases in their concentrations. For SP2 sample, approximately 50% of the algae was reduced with a 15 W UV-C radiation lamp within an hour. With 30 W UV-C lamp, up to 87% algae can be reduced in an hour. For the more UV-resistant SP1 samples, however, this can take more than 32 h regardless of being the same species (Fig. 1). Clearly, the reduction is dependent of the agent (i.e., variants of the same species). To validate the effectiveness and potential application of our treatment system, we constructed a field chamber using four 15 W UV-C bulbs in a 0.6×2 m chamber at the Beijing Botanical Garden in July of 2016. For the treatments, the field chamber was filled with lake water and observations were recorded every 2 h for 12 h. Due to the lack of equipment, only true color photos were taken to see the effects of the UV-C treatment. After 12 h of treatment, it was observed that the water within the treatment system was clear. The observations were consistent with our experimental results (Figs. 1–3) throughout three replications. More importantly, we took one sample prior and one after 1 h of treatment to compare and examine cell structures under a
Fig. 4. Relationships between predicted algal concentration (Chl RFU) (1000×PC s ml−1) and measured values from an independent experiment of UV-C radiation treatments with different initial conditions using the SP1 samples.
overestimations seemed more pronounced when algal concentration was high. Inclusion of initial density into the model, based on the empirical relationships in Fig. 3a will be a necessity to improve the accuracies of the models.
4. Discussion With increasing and more widespread of algal blooms in inland water systems, one of the forefront needs is the clear scientific basis when faced with public panic and institutional pressure. One of the most urgent needs to address this pressing issue is to seek naturebased, effective solutions to reduce algal concentration. In order to maintain the quality of the water, it is safest to not resort to the use of
Fig. 5. Microscopic photos (true color) of natural water with algal blooms in a lake of the Beijing Botanical Garden before and after 1 h treatment with 15 W UV-C radiation. Algae cells appeared to be ruptured after the treatment. These photos were taken at 2 spatial resolutions to demonstrate the breakdown of algal cells.
187
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
of toxins without rupturing the cells (Ou et al., 2011). In regards to effectively applying our results in freshwater lakes, an apparatus can be built and installed onto floatation devices (e.g., boats, buoys). To increase the engineering design of a treatment system, one may also need to consider circulating the water so that UV-C radiation can produce the strongest effects by evenly striking all parts of the water. Following the NBS concept, the treatment system can be solarpower-based. For example, in our trial treatment system that was applied in the Beijing Botanical Garden, a volume of 0.24 m3 of water was successfully cleaned in an hour (Fig. 5). Over a 24-h period, 5.76 m3 of water can be cleaned. If this similar solar-powered system is installed onto various floatation devices over Lake Erie, a large amount of water can be hypothetically cleaned daily (57.6 m3 per day with 10 systems). Because the treatment system is low-cost and solar-powered, this suggests that the deployment of this system over bodies of freshwater – even large lakes such as Lake Erie – is achievable. It is also possible that a large treatment facility can be built, but our objective in this study was to provide a scientific base for designing small-scale, portable treatment systems.
microscope (Fig. 5). It appeared that the cells had been destructed as a result of the UV-C treatment. This indicates that the algae was killed within one hour, despite that the sample remained green. Although the green color had disappeared within 2–4 h of treatment (Figs. 1–3), it is most likely that the cells had already died prior to the discoloration. Although UV-C radiation is an effective method of reducing algal concentration, other variables such as temperature may affect its effectiveness. For example, in a study by Wong et al. (2015), it was discovered that increasing temperatures could counteract the effectiveness of radiation. Different levels of UV-C radiation treatments and temperatures on C. vulgaris were administered in the study and it was concluded that UV-C radiation intensity was independent of changes in temperature. Although UV-C radiation effectively reduced algal concentrations, increased temperatures led to an improved ability for damaged cells to recover from UVR stress. This could potentially pose an issue for UV treatment systems when lakes or rivers rise in temperatures due to the effects of global warming. The underline mechanisms for the differences remain to be explored with consideration of the cultivation history (e.g., the media) within species variation and other potential biological and chemical properties of the water (nutrients, pH level, etc). The fact that a different response from SP3 samples to the same UVC treatment (Fig. 2c) from SP1 and SP2 samples also stimulate another potential hypothesis about the role of species interactions (e.g., elevated competition under stress; Jeffries and Mills, 1990) in responding to the UV radiation. Some natural following up steps would be analyzing the quantitative contributions of interactive casual relationships by algal density, different species of algae from in situ water sources, and UV-C radiation intensity through analysis of variance in more robust experiments (e.g., large replications). Future efforts can be further enhanced by mechanistic exploration of altered species interactions, resource limitations, population dynamics, nutrient cycling, species-environment relationships under different UV-C conditions, etc. One of the concerns for applying our approach on algae-polluted water in the field is that nutrients will be returned to the water, which will continue promoting the growth of algal population. However, our treatment will reduce the algal concentration, temporarily disrupting the growth. When applied properly in time and space, the treatment can mitigate the magnitude and scale of algal blooms (Ouyang et al., 2017). Another uncertainty is the resistance of damaged cells as a result of UV treatment. Algae could potentially build up resistance to UV radiation (Karsten and Holzinger, 2014), which is demonstrated in algae-UV relationships in alpine soils. In addition, algae responses to UV is also in condition to other variables (Wong et al., 2015). This suggests that future efforts are needed to examine the effects of confounding variables under varying environments from physiological and genetic perspectives (Ou et al., 2011; Brennan and Collins, 2015; McGivney et al., 2015; Sharma et al., 2015a, 2015b; Wong et al., 2015). A plausible finding of this study is the highly-reliable models that can be used for predicting the changes in algal concentration in the water under different UV-C radiation intensity and initial algal concentration (Figs. 1–3). The species we used in this study is a common species in many inland lakes (e.g., Lake Erie; Chaffin et al., 2014) and elsewhere (Ponnuswamy et al., 2013), suggesting that the model can be readily applied for treating the water from these lakes. Although these models were constructed for single species and in a controlled environment, we speculate the reduction rate will be at similar level for the natural lake water, where multiple species coexist. With this assumption, one can determine the amount of time for their treatment goals. However, our finding that the effects are densitydependent infers that treatment time needs to be set based on the algal concentration, with stronger UV-C radiation needed for high concentration water (Fig. 3). When studying other algal species in the future that may be toxic, such as Microcystis aeruginosa, it is imperative that we quantify an appropriate UV-C dosage that will degrade the release
5. Conclusions Both hypotheses regarding the effects of different UV-C radiation intensities on different algal concentrations and densities were accepted. Different intensities of UV-C radiation were applied to variations of Chlorella vulgaris. The 30 W UV-C radiation had a stronger effect than the 15 W UV-C radiation on reducing algal concentration. Another finding was that the UV-C radiations had the greatest impact on SP2 samples, and the lowest impact on SP1 samples. Furthermore, an independent study was conducted with only SP1 samples, but at different densities by diluting the samples with water. The results showed that the 30 W UV-C radiation still had the greatest impact and that as density increased, the reduction rates decreased. The independent experiment confirmed the density-dependent factor in the previous study. For each treatment, as well as UV-C radiation intensities and sample sources, empirical models were developed with high confidences (R2 > 84%). Using the information from the research and other similar future references, possible treatment systems could potentially be constructed and implemented indoors as well as in the fields. Acknowledgements We would like to thank many individuals for their advice, support and time for this project. Carol Stepien and Chris Mayer provided the essential equipment, including aquarium fish tanks, sets of fluorescent lights and air pumps to cultivate the algal growth; Wei Liao of the Michigan State University supplied a large sample of algae from his lab; Dr. Shenghua Gao and Xudong Zhang helped to construct a field pilot treatment chamber to validate the in situ application of the method at the Beijing Botanical Garden; Yao Cheng of the Chinese Academy of Sciences assisted in operating the microscope at IBCAS for changes at cell level. Two anonymous reviewers and the handling editor provided constructive suggestions to improved the quality of this paper. References Borderie, F., Alaoui-Sehmer, L., Bousta, F., Alaoui-Sossé, B., 2014. Cellular and molecular damage caused by high UV-C irradiation of the cave-harvested green alga Chlorella minutissima: implications for cave management. Int. Biodeterior. Biodegrad. 93, 118–130. Brennan, G., Collins, S., 2015. Growth responses of a green alga to multiple environmental divers. Nat. Clim. Change. http://dx.doi.org/10.1038/ NCLIMATE2682. Chaffin, J.D., Bridgeman, T.B., Bade, D.L., Mobilian, C.N., 2014. Summer phytoplankton nutrient limitation in Maumee Bay of Lake Erie during high-flow and low-flow years. J. Gt. Lakes Res. 40 (3), 524–531. Chen, R., Li, R., Dietz, L., Liu, Y., Stevenson, J., Liao, W., 2012. Freshwater algal
188
Environmental Research 156 (2017) 183–189
E.S. Chen, T.B. Bridgeman
Conserv. Lett.. http://dx.doi.org/10.1111/conl.12216. McGivney, E., Carlsson, M., Gustafsson, J.P., Gorokhova, E., 2015. Effects of UV-C and vacuum-UV TiO2 advanced oxidation processes on the acute mortality of microalgae. Photochem. Photobiol. 91, 1142–1149. Michalak, A.M., Anderson, E.J., Beletsky, D., Boland, S., Bosch, N.S., Bridgeman, T.B., DePinto, J.V., 2013. Record-setting algal bloom in Lake Erie caused by agricultural and meteorological trends consistent with expected future conditions. Proc. Natl. Acad. Sci. USA 110 (16), 6448–6452. Ou, H., Gao, N., Deng, Y., Qiao, J., Wang, H., 2011. Immediate and long-term impacts of UV-C irradiation on photosynthetic capacity, survival and microcystin-LR release risk of Microcystis aeruginosa. Water Res. 46 (2012), 1241–1250. Ouyang, Z., Shao, C., Chu, H., Becker, R., Bridgeman, T., Stepien, C.A., John, R., Chen, J., 2017. Algal bloom reduced carbon emission to the atmosphere in western Lake Erie. Remote Sens. 9 (1), 44. http://dx.doi.org/10.3390/rs9010044. Pogge, J., 2015. Anthropogenic Drivers of Harmful Algal Blooms, Specifically Microcystin, from Fresh to Marine Environments. Capstone, Nova Southeast University. Ponnuswamy, I., Madhavan, S., Shabudeen, S., 2013. Isolation and characterization of green microalgae for carbon sequestration, waste water treatment and bio-fuel production. Int. J. Bio-Sci. Bio-Technol. 5, 17–26. Sharma, K.K., Li, Y., Schenk, P.M., 2015b. Rapid lipid induction in Chlorella sp. by UV-C radiation. Bioenergy Res. 8, 1824–1830. Sharma, K.K., Ahmed, F., Schenk, P.M., Li, Y., 2015a. UV-C mediated rapid carotenoid induction and settling performance of Dunaliella salina and Haematococcus pluvialis. Biotechnol. Bioeng. 112, 2106–2114. Takeguchi, M., Oyobe, I., 1983. U.S. Patent No. 4,390,432. Washington, DC, U.S. Patent and Trademark Office. Wong, C.-Y., Teoh, M.-L., Phang, S.-M., Lim, P.-E., 2015. Interactive effects of temperature and UV radiation on photosynthesis of Chlorella strains from polar, temperate, and tropical environments: differential impacts on damage and repair. PLoS One 10 (10), e0139469. http://dx.doi.org/10.1371/journal.pone.0139469. Zelitch, I., 1971. Photosynthesis, Photorespiration and Plant Productivity. Academic Press, London, UK, 275.
cultivation with animal waste for nutrient removal and biomass accumulation. J. Biomass Bioenergy 39, 128–138. Coppen, S., 2015. Drivers of Harmful Algal Blooms in the North Sea (M.S. thesis). University of Gent. Hall, K.K., Giannetta, E.T., Getchell-White, S.I., Durbin, N.J., Farr, B.M., 2003. Ultraviolet light disinfection of hospital water for preventing nosocomial Legionella infection: a 13-year follow-up. Infect. Control Hosp. Epidemiol. 24 (8), 580–583. Hallegraeff, G.M., 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32 (2), 79–99. Heisler, J., Glibert, P.M., Burkholder, J.M., Anderson, D.M., Cochlan, W., Dennison, W.C., Dortch, Q., Gobler, C.J., Heil, 1, C.A., Humphries, E., Lewitus, A., Magnien, R., Marshall, H.G., Sellner, K., Stockwell, D.A., Stoecker, D.K., Suddleson, M., 2008. Eutrophication and harmful algal blooms: a scientific consensus. Harmful Algae 8 (1), 1–13. Ho, J.C., Michalak, A.M., 2015. Challenges in tracking harmful algal blooms: a synthesis of evidence from Lake Erie. J. Gt. Lakes Res. 41, 317–325. Jeffries, M., Mills, D., 1990. Freshwater Ecology: Principles and Applications. Belhaven Press, London. Jetoo, S., Grover, V.I., Krantzerg, G., 2015. The Toledo drinking water advisory: suggested application of the water safety planning approach. Sustainability 7 (8), 9787–9808. Kaplan, A., Badger, M.R., Berry, J.A., 1980. Photosynthesis and the intracellular inorganic carbon pool in the bluegreen alga Anabaena variabilis: response to external CO2 concentration. Planta 149 (3), 219–226. Karsten, U., Holzinger, A., 2014. Green algae in alpine biological soil crust communities: acclimation strategies against ultraviolet radiation and dehydration. Biodivers. Conserv. 23, 1845–1858. Loftin, K.A., Clark, J.M., Journey, C.A., Kolpin, D.W., Van Metre, P.C., Bradley, P.M., 2016. Spatial and temporal variation in microcystins occurrence in wadeable streams in the southeastern USA. Environ. Toxicol. Chem. 35. http://dx.doi.org/10.1002/ etc.3391. Lorenzen, C.J., 1966. A method for the continuous monitoring of in-vivo chlorophyll concentration. Deep Sea Res. 13, 223–227. Maes, J., Jacobs, S., 2015. Nature-based solutions for Europe's sustainable development.
189