• The distributions of dominant species of fish, phytoplankton, and benthic macroinvertebrates in the Pyeonggang Stream

• Potential effects of variations in water temperature on the ecosystem after the application of hydrothermal energy

• A plan for minimizing the influence of the water environment on hydrothermal energy supply operations.

Climate change has caused uncertainties in the supply of water and predictions of its consumption because of alterations in hydrological conditions, such as rainfall intensity. Increased energy consumption due to air-conditioning and rising temperatures have increased the economic burden of food crop production (Jung, 2018). New renewable energy solutions have been developed to deal with climate change. Examples of water-linked energy production systems are floating solar panels, offshore wind power, and geothermal energy.

Hydrothermal energy using temperature difference in Korea was first applied at the Mapo electric substation, which was used to heat buildings by collecting heat emitted from transformers on the ground (KIER, 2005). Additional advantages of cooling towers are reductions in noise and vibrations, the prevention of legionella, decreased costs of chemicals, and so on (Kim, 2020). Nationally, the utilization of water thermal energy contributes to the reduction of greenhouse gas emissions and benefits local economies by creating a new industry. In particular, changes in water temperature can decrease the amount of dissolved oxygen in rivers (Korea Environment Institute, 2014), the occurrence of eutrophication (Bates et al., 2008). Domestic cold-water fish inhabit good-quality water in the upper regions of rivers. They are protected as priority species because the populations of indigenous species are often low. Increasing water temperature changes the distribution of fish, and species that cannot move upstream become locally extinct (Allan, 1995). The geomorphological properties of freshwater fish depend on the temperature ranges that they can tolerate. In particular, because cold-water fish are sensitive, they should be moved upstream from water habitats that exceed their tolerance limits; otherwise, they may become extinct (Stefan et al., 2001).

Communities of benthic macroinvertebrates in river ecosystems have various and abundant compositions. In Korea, rivers are exposed to various disruptions and unstable water body substrate and the loss of surface substrates by precipitation creates unstable habitat environments for benthic macroinvertebrates (Boulton et al., 1992; Cobb et al., 1992). The life cycle of aquatic insects is highly influenced by temperature conditions. Variations in geological location affect the richness and habitation of benthic macroinvertebrates (Magnuson et al., 1979). In previous studies, benthic macroinvertebrates have been shown to be a good indicator organism for the evaluation of water environments because of their characteristics, such as a long-life cycle, community diversity, and ease of collection (Rosenberg and Resh, 1993; Shearer et al., 2015).

Phytoplankton is a primary producer that supports the energy and materials used in ecology systems (Keckeis et al., 2003). Phytoplankton reacts sensitively to variations in water environments. Therefore, detailed research on phytoplankton is required to determine variations in its water environment conditions. The mass propagation of phytoplankton causes several problems, such as the depletion of dissolved oxygen in water, and toxicity, as well as changes in species composition and dominant species (Xin et al., 2011). In the Nakdong River, water bloom by diatom typically occurs during the spring and cyanobacterial bloom during the summer (Joung et al., 2013). In nature, communities of phytoplankton are affected by temperature (Raven and Geider, 1988). The growth rate of cyanobacteria tends to increase, and floating control is facilitated by decreasing water viscosity as the temperature increases. Low light intensity can also control the dominance of cyanobacteria (Karl et al., 2003).

The West Nakdong River is controlled by two floodgates, the Daejeo floodgate upstream and the Noksan floodgate downstream, which were installed for agricultural use. The water mass is stagnant during most of the year in the West Nakdong River. Moreover, point pollutant sources such as sewage treatment plants and excreta treatment plants, as well as various nonpoint pollutant sources due to agriculture and livestock, are broadly distributed around the watershed of the West Nakdong.

In this study, the distributions of dominant species of fish, phytoplankton, and benthic macroinvertebrates were evaluated to assess variations in the water environment in the Pyeonggang Stream, which is part of the drainage system of the West Nackdong River. Potential effects of variations in water temperature on the ecosystem were simulated in two scenarios to prepare suitable plans for minimizing their influence by supplying hydrothermal energy to the Eco Delta Smart Village in Busan.

2.1. Research Area

The research area was bare land on which the construction of the Eco Delta Village began in 2019. Effluent is discharged into the Pyeonggang Stream from the Seobu sewage treatment plant of the Busan Environmental Corporation, which is located 7 km from the site. Evaluations of monitoring in this area were performed at three points: upstream of the Pyeonggang Stream (SW. 1); intake and discharge points in the Smart Village (SW. 2); and downstream of Pyeonggang Stream (SW.3). The three sampling points are presented in Fig. 1. Samples were taken in each of the four seasons, as follows: February 20 (winter); May 14 (spring); August 4 (summer); and October 12 (fall).

Figure 1. Sampling sites in the Pyeonggang Stream, Korea.

2.2. Water Analysis

Water was analyzed for items of pH, DO, BOD, COD, TOC, TN, TP, and SS. Total coliform, pH, and temperature were analyzed by a multiparameter YSI instrument (Pro Plus). The samples were preserved in an ice box before they were moved to the laboratory. Biochemical oxygen demand (BOD) was analyzed to determine the amount of oxygen consumed by the microbes at an incubation of 20℃ for five days. Chemical oxygen demand (COD) was analyzed using the potassium permanganate method. SS was filtered through a Watman GF/C and dried at 105-110℃. Dissolved organic carbon (DOC) was analyzed using a total organic analyzer (TOC- L, Shimadzu) after the samples were filtered at 0.45 μm and controlled to pH 2 with an HCl solution and quantified by measuring the amount of non-purgeable organic carbon (NPOC). The ascorbic acid method was applied at 880 nm to determine total phosphorus. The UV spectrophotometric method was applied at 220nm with alkaline potassium persulfate digestion at 120-124℃ to determine total nitrogen (TN). Lactose broth was used in the estimation of total coliform and incubated at 35±1℃ for 24±2 hr. Cultured solution taken from a positive tube with a loop was inoculated to be confirmed on the brilliant green lactose bile with an inoculation loop and incubated at 35±1℃ for 48±3 hr. The gas occurrence was shown to be positive.

Benthic macroinvertebrates were collected by a surber net (50 cm× 50 cm) for quantitative analysis repeated three times. In the qualitative analysis of the benthic macroinvertebrates, a hand net and hard bottom scraper were used. The samples were fixed in 70% alcohol and preserved in Kahle’s solution. The fish were collected by skimming nets (mesh size: 50 × 50 mm) and cast nets (mesh size: 50 × 50 mm). After identification, the fish were released on site. The fish were identified based on previous studies (Uchida, 1939; Jung, 1977; Kim, 1997; Choi et al., 2002). Taxonomy was according to Nelson (1994). The health of the water environment was assessed by the benthic macroinvertebrate index (BMI) (Kong et al., 2018).

The phytoplankton samples were analyzed using a Sedwick-Rafter chamber and enumerated by the Schoen method. Phytoplankton was identified using optical microscope at 400–1000 magnification. After phytoplankton was identified according to taxon, it was enumerated and calculated as cell number per mL. Phytoplankton was identified according to previous studies (Hirose and Yamagishi, 1977; Jung, 1993).

2.3. Evaluation by EFDC of Simulated Variations in Water Temperatures

Hydrothermal energy at environmental fluid dynamics code (EFDC), which is a three-dimensional hydrodynamic model developed by the Virginia Institute of Marine Science, was used to simulate variations in water temperature in this watershed. Variations in vertical layers were not considered. The sigma stretching coordinate system, which was divided into an equal number of layers, was applied because there is little variation in water depth in the West Nakdong River. Simulated sections were included for the West Nakdong River (Daejeo floodgate-Noksan floodgate), the Macdo River (Macdo pump station and the Sinpo pump station), the Joman River, and tributary rivers (the Yean Stream, Joojung Stream, Sineo Stream, Jisa Stream, and Pyeonggang Stream). Topographical data were combined with data on cross-section measurements in the master plan report on the West Nakdong River (2012). The mean size of the horizontal grid of the West Nakdong River was 37.7 m × 45.4 m, and the number of horizontal grids was 4,903. The mean orthogonality was 0.517°. The boundary conditions were as follows: West Nakdong River, five points; Macdo River, four points; Joman River, two points; Pyeonggang Stream, two points.

The revision period was from January 1, 2018 to December 31, 2018. The amounts of inflow and outflow were used with data from the Gangseo-gu office in Busan on the tank model simulation, in addition to K-water in major rivers and actual measurements of flow rates through daily floodgates and pump stations. Abundant water flow, ordinary water flow, low water flow, and mean drought water flow were measured at 45,868 m³/d, 20,399 m³/d, 12,403 m³/d and 3,968 m³/d, respectively, after performing a flow duration analysis of inflow rate data on the Pyeonggang Stream. Meteorological data were used as input data, and hourly data in 2018 were considered weighted values according to the distance from the meteorological observatory in the cities of Busan and Kimhae. Data on water temperature were used with daily temperature data collected in 2018 from the water environment information system.

3.1. Analysis of Water and Deposits

The mean water temperatures were 9.5℃, 20.9℃, 28℃, and 18.3℃ in February, May, August, and October, respectively (Table 1). DO and pH levels were the lowest, while COD was the highest during the summer. The mean amount of coliform during the summer was eight times higher than during the winter. Park et al. (2019) reported that water quality from 2008 to 2017 worsened in summer but improved in winter in the West Nackdong River.

Table 1 . Variations in water quality at three sampling points in the Pyeonggang Stream.

Items	Feb.			May			Aug.			Oct.
	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3
Temp.(℃)	9.9	9.4	9.2	21	21	20.8	28	28	28	18.3	18.3	18.4
pH	7.9	8.4	8.2	8.2	8.6	8.6	7.6	7.6	7.9	8	8.4	8.5
DO(mg/L)	7.9	8.6	9.9	11.3	10.3	10.3	6.7	7.7	8.8	6.7	8.1	8.3
COD(mg/L)	10.9	10.2	10.3	13.6	10.8	10.6	16.1	12.6	14.2	13.4	12	11.2
BOD(mg/L)	4.4	3.6	3.6	4.8	1.6	1.2	4	3.6	3.2	3.8	2.9	2.3
TOC(mg/L)	5	4.2	4.2	7	5	4.7	4.3	3.8	3.6	5.2	3.7	3.6
SS(mg/L)	20.0	25.0	27.0	36.5	19.0	19.0	32.0	26.5	28.0	28.5	25.0	22.5
T-N(mg/L)	4.3	4.1	4.2	3.4	3.7	2.5	3.1	2.9	2.6	2.7	3.6	4
T-P(mg/L)	0.186	0.1	0.102	0.318	0.112	0.095	0.158	0.203	0.145	0.132	0.088	0.076
Total coliforms(CFU/100mL)	540	240	490	350	540	790	920	540	540	100	79	70

The highest value of COD was 16.1 mg/L at SW 1. Based on COD, it corresponded to “very bad” in Korea’s life environmental standards. The highest levels of COD and BOD (4.8 mg/L) were found at SW1 throughout all seasons. SS were 24.0, 24.8, 28.8, and 25.3 mg/L in February, May, August, and October, respectively. Metals such as Cd and Hg. Lead (Pb) were not detected in these samples. Kang et al. (2013) reported slight differences in discharge in most watersheds of the West Nakdong River between the flood season and non-flood season on the variation of rainfall due to climate change because it is a stagnated area. They recommended that water quality could be improved by controlling the water velocity through floodgates.

In sediments at SW1, SW2, and SW3, the levels of Pb and As were not observed to be toxic to benthic macroinvertebrates (Table 2). However, 0.07 mg/kg of mercury was found at SW-1, and 46.7 mg/kg of Cu was found at SW-1. These levels could significantly affect benthic macroinvertebrates. The level of total phosphorus at SW-1 throughout all seasons was 1,954 mg/kg, and the highest value was 2,800 mg/kg, which has an influence on benthic macroinvertebrates. The highest level of total nitrogen at SW-1 was 5,560 mg/kg. PCBs were not found in the sediments.

Table 2 . Analysis of deposits in the Pyeonggang Stream at three sampling points.

Items	Feb.			May			Aug.			Oct.
	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3
Ignition loss(%)	5.7	3.1	7.7	9.8	8.4	4	3	3.8	6.2	2.9	3.4	6.5
COD(mg/kg)	1.3	0.36	1.37	1.77	0.64	0.52	1.38	0.32	0.62	1.64	0.52	0.5
T-N(mg/kg)	3,002	596	2,974	5,560	2,276	1,536	3,372	702	2,745	4,058	2,606	1,246
T-P(mg/kg)	2,800	557	975	1,957	954	698	1,770	518	934	1,289	720	621
Cu(mg/kg)	23.6	25.2	17.9	40.7	37.2	33.3	46.7	35.3	37	11.3	21.9	33.6
As(mg/kg)	5.01	4.04	3.84	3.99	2.97	2.67	4.06	4.08	2.82	5.09	4.82	6.8
Hg(mg/kg)	0.07	0.03	0.04	0.06	0.04	0.04	ND	ND	ND	ND	ND	ND
Pb(mg/kg)	22	24.6	20.9	26	30.8	30.4	36.1	36.5	35.5	ND	10	18.1
Zn(mg/kg)	168.6	148.9	136.8	204.9	179.4	170.1	201.7	179.8	181.3	244.9	218.6	254.4
F(mg/kg)	222	271	277	54	115	68	177	167	198	178	192	182

3.2. Species Composition of Fish

Fish fauna and their populations were poor in the research area. During the survey periods, five species, Carassius auratus, Zacco platypus, Micropterus salmoides, Lepomis macrochirus, and Tridentiger brevispinis, were collected (Table 3). The predominant species in May, July, and October were Carassius auratus, Lepomis macrochirus, and Zacco platypus respectively. Lepomis macrochirus and Micropterus salmoides are invasive species. Tridentiger brevispinis is distributed in Korea and Japan. Tridentiger brevispinis was found only in July. Zacco platypus was observed only in October.

Table 3 . Seasonal variations in fish fauna and composition at three sampling points.

Fish	Feb.			May			Jul.			Oct.
	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3	SW-1	SW-2	SW-3
Carassius auratus	-	-	-	12	8	9	5	2	3	3	3	2
Zacco platypus	-	-	-	-	-	-	-	-	-	-	5	12
Micropterus salmoides	-	-	-	1	-	-	1	-	-	2	-	9
Lepomis macrochirus	-	-	-	8	-	-	24	11	9	-	2	3
Tridentiger brevispinis	-	-	-	-	-	-	4	2	-	-	-	-
Total species	-	-	-	3	1	1	4	3	2	2	3	4
Populations	-	-	-	21	8	9	34	15	12	5	10	26

The predominant species is Lepomis macrochirus, which prefers to live in a lentic zone, and the subdominant species is Carassius auratus, which is a tolerant species distributed in spring, summer, and fall, except winter. Micropterus salmoides is carnivorous and eats various species such as amphibians and reptiles as well as fish and aquatic insects (Lee et al. 2009). Lee et al. reported Micropterus salmoide intake zooplankton such as copepods as well as fish, aquatic insects and benthic macroinvertebrates. Lepomis macrochirus originated in North America, where it is known to be an omnivorous species. After being introduced in Korea, it became carnivorous and insectivorous (Byon and Jeon, 1997). In this study no fish were caught during the winter season.

Seventeen individuals of Zacco platypus were found in the research area. Zacco platypus is distributed on a wide habitat from oligotrophic to eutrophic conditions (Jeon, 1980; Kim,1997). It is tolerant of water pollution and artificial environmental variations, such as dam construction, weir installment, and aggregate collection (Jeon, 1980). Ko et al. (2012) reported that eutrophication conditions in lakes create a suitable environment for Zacco platypus. When environmental conditions deteriorate, Zacco platypus feed on organic matter as well as aquatic insects on gravel and sand or periphyton (Kim and Kim 1975; Kim, 1997; Kim et al., 2010). In this area, concentrations of phosphorus and nitrogen in water and sediment are high. Zacco platypus was predominant in October. Zacco platypus concentrically inhabits places where sand is silted up and water depth is shallow. Lepomis macrochirus is distributed in deep water and stagnant water flow in the downstream region of Gyeongan Stream (Lee et al., 2013).

Eighteen species were previously reported from 2015 to 2019 (Busan Metropolitan Corporation et al., 2020; Busan Metropolitan City, 2018), as follows: Cyprinus carpio, Erythroculter erythropterus, Hemiculter eigenmanni, Opsariichthys uncirostris amurensi, Acheilognathus macropterus, Carassius cuvieri, Mugil cephalus, Silurus asotus Hemibarbus labeo, Coilia nasus, Plecoglossus altivelis, Acheilognathus majusculus, Leiocassis nitidus, Rhinogobius brunneus, and Tridentiger brevispinis. However, these reports did not include Zacco platypus. Ten species were found from 2011 to 2013 (Busan Metropolitan Corporation et al., 2013), and Squalidus chankaensis tsuchigae were found at the site. In 2008, The Ministry of Environment in Korea reported 20 species of freshwater fish in this area. The following species were included in the list. Zacco platypus, Squalidus gracilis majimae, Oryzias latipes, Hyporhamphus sajori, Rhodeus ocellatus, Pseudorasbora parva, Gasterosteus aculeatus, Acanthogobius lactipes, and Rhinogobius giurinus. Among these species, Acheilognathus majusculus and Squalidus chankaensis tsuchigae were endemic. Neither was observed after 2018. In particular, Acheilognathus majusculus has been designated an endangered species by the Ministry of Environment in Korea since 2017.

Kang et al. (2013) analyzed increases in water temperature at 0.69 ℃, 1.76 ℃, and 2.32 ℃, finding that they affected fish habitat by 21.9%, 36.3%, and 51.4%, respectively, in 22 fish species in the Nakdong River. Rhodeus uyekii showed the highest maximum thermal tolerance (33.1 ℃), followed by Acheilognathus macropterus, Zacco platypus, and Misgurnus anguillicaudatus, which had a relatively high maximum thermal tolerance temperature at 31℃ (Kang et al., 2013). Therefore, during the operation of hydrothermal energy supply facilities, the condition of onsite water temperatures should be controlled to maintain the thermal tolerance temperatures of fish species.

3.3. Species Composition of Benthic Macroinvertebrates

Twenty-five species of benthic macroinvertebrates belong to 21 families, 13 orders, and seven classes (Table 4). The classes included the following: one species of Platyhelminthes (4.0%); seven species of Mollusca (28.0%); two species of Annelida (8.0%); and 15 species of Arthropoda (60.0%). The ratio of non-insectivores was higher. The highest number of species (16 species) was observed in July. However, the highest population of benthic macroinvertebrates was observed in October (228 individuals). In the research area, 728 individuals were observed. The dominant species was Chironomidae sp., and the subdominant species was Caridina denticulata denticulate. Chironomidae sp. is highly tolerant of variations in water environments. The number of Chironomidae sp. was the highest during the winter season, which was 67.5% compared with the total population.

Table 4 . Seasonal variations in number of benthic macroinvertebrates in the Pyeonggang Stream.

Species	Feb.	May	July	Oct.
Dugesia japonica Ichikawz & Kawakatsu	2		9
Pomacea camaliculata		11	9	10
Cipangopaludina chinensis malleata (Reeve)	1	1	2	1
Lymnaea auricularia (Linnaeus)	2	43	28	11
Physa acuta Draparnaud	11	9	7	9
Gyraulus chinensis (Dunker)		7	2	5
Hippeutis cantori (Benson)	2	2		3
Anodonta woodiana (Lea)	3
Limnodrilus gotoi Hatai	4	5	10
Erpobdella lineata (Muller)	1	6	3
Gnorimosphaeroma naktongense Kwon & Kim				7
Asellus hilgendorfii Bovalius	2	3	3
Caridina denticulata denticulata (De Haan)	15	21	26	79
Palaemon paucidens (De Haan)		8	18	39
Macrobrachium nipponense (De Haan)			4
Ecdyonurus levis (Navás)				6
Calopteryx japonica Selys				8
Paracercion clamorum (Ris)				11
Paracercion hieroglyphicum			5
Davidius lunatus (Bartenef)			19
Appasus japonicus Vuillefroy			45
Appasus major (Esaki)		13
Aquaris paludum (Fabricius)	6	11	6	12
Chironomidae sp.	102	15	20	22
Diversity (H’)	1.28	2.3	2.37	2.17
Dominance (DI)	0.77	0.41	0.38	0.52
Richness (RI)	2.19	2.58	2.85	2.58
BMI	26.1	38.0	43.8	54.1

During the winter season, the index of dominance was 0.7, 0.97, and 0.94 at SW-1, SW-2, and SW-3, respectively. The level of richness index was the highest at SW-1 in all seasons. During the spring and summer, Lymnaea auricularia was dominant and subdominant, respectively, showing resistance to stress conditions in the water. During the fall and winter, Caridina denticulata was dominant and subdominant, respectively. Caridina denticulata was distributed mainly at SW-1 in the Pyeonggang Stream. This area was observed to be inhabited by highly resistant groups and extreme occurrences of specific species. This was because excessive amounts of nutrients, such as nitrogen and phosphorus, are maintained in this area by various physical disturbances, such as construction work and vehicle movement.

The Ephemeroptera, Plecoptera, and Trichoptera (EPT) group is sensitive to variations in the water environment (Lenat, 1988). Only one species, Ecdyonurus levis, was observed. Seven species of the Odonata, Coleoptera, and Hemipter group (OCH) were observed in the lentic zone surrounded by vegetation. Cristaria plicata, which is in the list of endangered wild animals in Korea, were not found in the present study area, although they were observed in 2018 (Busan Metropolitan Corporation et al., 2020).

Aquatic insects, which generally inhabit over 80% of a stream, were observed at 40% in the study area. In Korea, over 1,500 species of aquatic insects were recorded (Jung et al., 2020), which have an important role at the trophic level. Decreases in the EPT group and observation of the Chironomidae family, Physa acuta and Limnodrilus gotoi, indicated water deterioration. Because of artificial disturbances, there was little variety in aquatic species.

BMI was the lowest (18.0) at SW. 3 in February. The highest BMI (58.3) was at SW. 2 in October. The mean BMI values were 26.1, 38.0, 43.8, and 54.1 in February, May, July, and October, respectively. These values indicated that the health of the water ecology system was “bad.” In February, the values were ranked E grade under BMI 35. In May and July, the water quality by BMI was “bad” (D level).

According to the living environmental standards in Korea, the COD of water quality was VI (very bad), and the mean BMI (40.5) in all seasons was D (bad, 35 ≥ BMI < 50) in the Pyeonggang Stream. These results indicate that the health of the ecosystem and water quality were in a deteriorated condition. Therefore, water velocity should be secured because this area is stagnant; moreover, the effects of nonpoint pollutant sources on the ecosystem should be controlled. Yoon et al. (2013) reported that because the West Nackdong River is a stagnated watershed, water quality should be improved by water velocity management and floodgate control.

3.4. Species Composition of Phytoplankton

Phytoplankton is a crucial indicator of variations in water environments because it is sensitive to changes in these environments. Detailed conditions of phytoplankton should be determined to predict future variations in specific water bodies due to increases in the water temperature. In this survey, a total of 82 species, 35 families, and 22 orders, were verified in the area. In February, May, July, and October, 53, 52, 44, and 60 species of periphyton, respectively, were analyzed. Scenedesmus acuminatus, Scenedesmus quadricauda, Cosmarium sp., Cyclotella meneghiniana, Fragilaria sp., Synedra acus, Cocconeis pediculus, Cocconeis placentula, Cymbella sp., Cymbella tumida, Cymbella turgidula, Navicula capitata, Navicula cryptocephala, Navicula pupula were commonly observed in all seasons.

Commonly observed classes of phytoplankton were Chlorophyceae and Bacillariophyceae. Bacillariophyceae were found to be at 84.9%, 71.2% 68.2%, and 59.0% in February, May, July, and October, respectively, compared with all identified species (Fig. 2). Phytoplankton species identified according to season were in the following classes: Bacillariophyceae > Chlorophyceae > Cyanophyceae during the winter; Bacillariophyceae > Chlorophyceae > Cyanophyceae > Chrysophyceae during the spring and summer; Bacillariophyceae > Chlorophyceae > Cyanophyceae > Euglenoidea > Chrysophyceae during the fall. Bacillariophyceae was predominant for 300 days downstream of the Nackdong River (Son, 2013a).

Figure 2. Composition of phytoplankton species according to monthly variations.

The order of phytoplankton observed for standing crops of periphyton was as follows: Cyanophyceae > Bacillariophyceae > Chlorophyceae > Chrysophyceae during the summer and Cyanophyceae > Bacillariophyceae > Chlorophyceae > Euglenoidea > Chrysophyce during the fall for standing crops of periphyton (Fig. 3). During the spring and winter, the same orders of standing crops as of the identified species were observed. In the Nackdong River system, the amounts of standing crop of phytoplankton were higher downstream compared with midstream (Son, 2013b).

Figure 3. Seasonal variations in standing crops of phytoplankton in the Pyeonggang Stream.

During the winter season, the dominant species was Aulacoseira ambigua, and the subdominant species was Cyclotella meneghiniana. Cyclotella meneghiniana appeared to be the highest standing crop at SW. 2 in February (208 cells/mL). In May, the dominant species was Nitzschia palea, and the subdominant species was Aulacoseira ambigua (Table 5). The highest value of Nitzschia palea (416 cells/mL) was recorded at SW. 3. in May. In July, the dominant genus was Microcystis sp. and the subdominant genus was Aphanizomenon sp. in July. The highest values of Microcystis sp. (11,184 cells/mL) were recorded at SW. 3 in July. Dominant species and subdominant species were Microcystis sp.1 and Oscillatoria sp., respectively, in October. The highest value of Microcystis sp.1 was recorded at SW. 3 in October. The optimal temperatures for Nitzschia palea are between 15℃ and 25℃ (Kwon et al., 2011) and between 28℃ and 32℃ for Microcystis sp. 1 (Nalewajko and Murphy, 2001).

Table 5 . Dominant species and standing crops of phytoplankton in the Pyeonggang Stream.

Classification	Feb		May		July		Oct.
	Dominant species	Standing crops (cell/mL)	Dominant species	Standing crops (cell/mL)	Dominant species	Standing crops (cell/mL)	Dominant species	Standing crops (cell/mL)
Bacillariophyceae	Aulacoseira ambigua	493	Nitzschia palea	1097	Aulacoseira granulata	778	Navicula sp.	2518
	Aulacoseira sp.	256	Nitzschia sp.	409	Cyclotella meneghiniana	804
	Melosira varians	288	Aulacoseira ambigua	518
	Cyclotella meneghiniana	354
	Fragilaria capucina	208
Cyanophyceae					Anabaena sp.	2083	Anabaena sp.	2003
					Aphanizomenon sp.	3288	Microcystis sp.1	11835
					Microcystis sp.	26972	Microcystis sp.	2 2966
							Oscillatoria sp.	4784
Chlorophyceae			Scenedesmus acuminatus	450
			Scenedesmus quadricauda	419
Diversity (H’)		3.33		3.23		0.81		2.19
Dominance (DI)		0.21		0.27		1.26		0.56
Richness (RI)		6.28		5.87		0.35		5.72

Microcystis sp. are representative blooms in Korean waters. These are dominant species during summer in the Nakdong River because they adapt well to strong sunlight and are well-positioned for photosynthesis (Yu et al. 2014). Xia et al. (2011) reported that water blooms by Microcystis sp. occurred over 25 ℃ and that maximum growth occurred between 28 ℃ and 32 ℃. Increases in water temperature near discharge points in the Pyeonggang Stream may extend periods of water bloom because the optimal growth of Cyanophyceae occurs mainly over 25 ℃.

The phytoplankton biomass is proportional to nutrient concentrations; limited nutrients influence the growth of phytoplankton (Heck and Kilham, 1988). Son (2013) showed that phosphorous limited the growth of phytoplankton downstream of the Nackdong River. Smith reported TN/ TP was over 17, and phosphorous could be a limiting factor in the growth of phytoplankton (Smith, 1982). Smith (1983) found that Chlorophyceae were predominant when TN:TP was less than 29:1, based on data collected in the temperate region.

However, the findings of the present study showed that Cyanophyceae dominant during the summer and fall seasons. The ratios of TN and TP were 19.6, 14.3, 17.9 at SW. 1, SW. 2, and SW. 3 during the summer season, respectively. and Cyanophyceae occupied 86.9%. The ratios of TN and TP were 20.5, 40.9, and 52.6 at SW. 1, SW. 2, and SW. 3, and Cyanophyceae was 76.9% during the fall season. However, Yu et al. (2015) found that the ratios of TN and TP in the Nakdong River were lower than 29 in early stage of algae blooms in water. Lee et al. (2002) showed that variations in standing crops of phytoplankton were unrelated to variations in TN and TP in the Nackdong River. The reason is that phytoplankton growth is not sensitive to changes in nitrogen and phosphorus levels under nutrientrich conditions (Lee et al., 2002).

The Shannon Diversity Index (H') showed 3.33, 3.23, 0.81, and 2.19 in February, May, July, and October, respectively. The Dominance Index (DI) showed 0.21, 0.27, 1.26, and 0.56 in February, May, July, and October, respectively. The differences in index values were significant during the seasonal variations. The DI had the highest value (1.42) at St. 1 in the summer and the lowest value (0.20) at St. 1 in the winter.

3.5.Water Temperature Variations due to River Water Hydrothermal Energy

In Scenario 1 (quantity of water intake 1,600 m³/day, water temp. difference; ±5℃), the applications of hydrothermal energy at DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100 (downstream 100 m, downstream 50 m, downstream 20 m, downstream 10 m, upstream 1 0 upstream 20 m, upstream 50 m, upstream 100 m, respectively, from the discharge point) were simulated to observe variations in water temperature (Fig. 4). Monthly mean temperatures ranged as follows: -0.5~0.34℃, -0.73~0.49℃, -1.24~0.67℃, -1.53~0.93℃, -1.49~0.52℃, -0.99~0.36℃, -0.74~0.30℃, -0.55~0.30℃ for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively (Appendix 1). Maximum increases in daily water temperature was predicted to be 1.06℃, 1.07℃, 1.27℃, 2.32℃, 1.81℃, 1.87℃, 1.75℃, 1.59℃ for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively. Maximum decreases in daily water temperatures were expected to be -1.69℃, -2.79℃, -3.33℃, -3.85℃, -4.27℃, -2.44℃, -1.90℃, and -1.44℃ for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively (Table 6).

Table 6 . Maximum values of daily water temperatures predicted due to river water hydrothermal energy in Scenario 1 and 2 at different points.

Water Temp.	Scenario 1								Scenario 2
	DNS100	DNS50	DNS20	DNS10	UPS10	UPS20	UPS50	UPS100	DNS100	DNS50	DNS20	DNS10	UPS10	UPS20	UPS50	UPS100
Maximum Increase	1.06	1.07	1.27	2.32	1.81	1.87	1.75	1.59	0.48	0.56	0.79	0.96	1.35	1.08	0.70	0.44
Maximum Decrease	-1.69	-2.79	-3.33	-3.85	-4.27	-2.44	-1.90	-1.44	-1.24	-1.62	-2.24	-2.56	-1.76	-1.42	-1.14	-1.11

Figure 4. Comparison of variations in water temperature currently and in Scenario 1 at different points. *Date of maximum water temperature differences during the year (November 24 and October 27).

Based on the point of DNS 10, UPS 10, mean values of temperature differences were predicted to be between -1.10℃ and -1.15℃ from November to April, when residential heating is the highest. The mean values of temperature differences were predicted to be between 0.63℃ and 0.48℃ from May to October, when airconditioning increased in Scenario 1.

Variations in water temperature in the Pyeonggang Stream due to hydrothermal energy production can affect the water ecosystem. In the Eco Delta Smart Village, stream intake quantity was planned to be between 1,600 m³/d and 2,800 m³/d to supply energy to the model housing complex. However, intake volume could be maintained under 1.600 m³/day with the complementary operation of geothermal and solar heat sources. Currently, both systems are planned to supply energy to the model housing complex in the Eco Delta Smart Village.

The following countermeasures are recommended to minimalize the environmental effects on animals and plants in surface water caused by variations in water quantity and temperature during the intake and drainage of stream water. The location of the discharge point should be located downstream over 150 m from the intake point. This will avoid the re-intake of return flow. The second recommendation is the addition of as much stream water in the perforated pipes as the intake quantity before discharge at the outlet. These are discharged through the embedded rock layer to extend the retention time of the discharge water to recover the temperature of the water (Fig. 5). In Scenario 2, 1,600 m³/d of dilution water were added, and a discharge point was designated at points 150m downstream from the intake points and then released through perforated pipes into the Pyeonggan Stream based on the same operating condition as in Scenario 1 (i.e., quantity of water intake at 1,600 m³/day and water temperature differences ±5℃).

Figure 5. Recommendations for reducing variations in water temperature (The design drawing in the background of Fig. 5 was originally supplied by K-Water).

Monthly mean temperatures ranged as follows: -0.35 ~ 0.23, -0.43 ~ 0.28℃, -0.63 ~ 0.45℃, -0.74–0.53℃, -0.57 -0.47℃, -0.46–0.33℃, -0.42–0.23℃, -0.37–0.13℃ for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively (Appendix 1 and Fig. 6). Maximum increases in daily water temperature were assessed as follows: 0.48℃, 0.56℃, 0.79℃, 0.96℃, 1.35℃, 1.08℃, 0.70℃, 0.44℃ for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively. Maximum decreases in daily water temperatures were assessed as follows: -1.24℃, -1.62℃, -2.24℃, -2.56℃, -1.76℃, -1.42℃, -1.14℃, and -1.11℃ for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively. Based on the points of DNS 10 and UPS 10, the mean values of temperature differences were predicted to be between -0.51℃ and -0.47℃ from December to April, with heating. The mean values of temperature differences were predicted to be between 0.44℃ and 0.28℃ from May to October, with air-conditioning in Scenario 2.

Figure 6. Comparison of variations in water temperature currently and in Scenario 2 at different points. *Date of maximum water temperature difference of the year (December 7 and May 22).

Recovery times were assessed as follows: 5 hr, 6 hr, 7 hr, 12 hr, 23 hr, 23 hr, 23 hr and 23 hr during cooling at DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100, respectively (Table 7). In all areas observed in this study, high temperatures decreased within one day in the simulation. As the distance from the discharge points increased, the recovery time decreased.

Table 7 . Water Temperature Recovery Times for DNS 100, DNS 50, DNS 20, DNS 10, UPS 10, UPS 20, UPS 50, and UPS 100 in the Pyeonggang Stream in Scenario 2.

Points	Recovery time (hr)
	Heating	Cooling
DNS 100	19	5
DNS 50	19	6
DNS 20	19	7
DNS 10	19	12
UPS 10	10	23
UPS 20	9	23
UPS 50	9	23
UPS 100	9	23

This study was performed to determine changes in the water ecosystem of the Pyeonggang Stream and evaluate the possible effects of hydrothermal energy on the drainage system of the West Nackdong River caused by cooling and heating in the Eco Delta Smart Village in Busan. The status of fish, benthic macroinvertebrates, phytoplankton inhabited, and water quality in the Pyeonggan Stream, which is reservoir-like and controlled by two water gates, were monitored for the effects of water temperature on species distribution and dominant species.

The predominant species of fish was Lepomis macrochirus, and the subdominant species was Carassius auratus, which is resistant to water pollution. The distribution of fish species was poor in this area. The mean BMI value was 40.5, which indicated that environmental health was bad (D), and the water quality assessed by COD was very bad. The long-term monitoring of the environment and ecology system is required when the hydrothermal energy station is in operation to minimize its effects on the environment and aquatic life in the water ecology system.

The percentages of aquatic insects were low. Among benthic macroinvertebrates, aquatic insects were as correspond to 40% in the stream. The predominant species of benthic macroinvertebrates were Chironomidae sp., Lymnaea auricularia, Appasus japonicus, and Caridina denticulata denticulata in February, May, July, and October, respectively. Bacillariophyceae during spring and winter and Cyanophyceae during summer and fall were typical.

Variations in water temperature were found at -0.19℃ and 0.59℃ during cooling (from May to October) and heating (from November to April), respectively, at a point 10m downstream from the discharge points. Water temperatures assessed at -0.20℃ and 0.68℃ during cooling and heating, respectively, at a point 10 m upstream. These temperatures were assessed for the condition after operating at the hydrothermal energy facility, which will supply heating and cooling to the model housing complex. The following operating factors are applied: quantity of water intake and dilution water, 1,600 m³/d and differences in water temperature of ±5℃. Recommendations for reducing variations in water temperature include the installation of discharge points to locations 150 m away from intake points. In addition, equal quantities of intake water should be added before releasing discharge through the embedded rock layer.