Characteristics and Changes in Water Quality Based on Climate and Hydrology Effects in the Cirata Reservoir

Abstract

This research aimed to identify water quality changes in the Cirata Reservoir and the factors affecting them in terms of hydrology and climate. The sampling was carried out in both the rainy and dry seasons at 12 locations in the Cirata Reservoir. The Mann–Whitney U-test (different test) results showed that salinity, total suspended solids (TSS), the potential of hydrogen (pH), nitrate (NO₃-N), phosphate (PO₄), nitrate and phosphate content in the sediment were significantly different (α < 0.05) between the rainy and dry seasons. The principal component analysis (PCA) results showed that the water quality characteristics in the Cirata Reservoir in the dry season were influenced by environmental conditions in the reservoir, especially by the floating cage aquaculture and climate conditions. The high solar radiation, low rainfall, and floating cage aquaculture increased the pH and amounts of dissolved oxygen (DO), ammonia (NH₃-N), PO₄, nitrate and phosphate in the sediment while decreasing transparency, salinity, TSS, and NH₃-N. During the rainy season, the high runoff from Citarum Watershed controlled the water quality characteristics of the Cirata Reservoir. In this season, transparency, salinity, pH, DO, NH₃-N, PO₄, nitrate and phosphate in the sediment increased, while TSS tended to be low. In general, the water volume addition decreased the nutrition and salinity concentration in the water body. However, a distinct phenomenon occurred in the Cirata Reservoir. The runoff from agriculture, settlement, livestock, and the Citatah Karst in the Upper Citarum Watershed increased nutrition and salinity in the reservoir. Land use in the Citarum Watershed and floating cage aquaculture had an important role in the reservoir water quality.

1. Introduction and Background

The maintenance and improvement of water quality in lentic systems is a challenge faced by the entire world. In particular, for artificial environments such as reservoirs, the water quality is affected by natural phenomena and human activities in the vicinity [1]. Changes in climate, season, and rainfall intensity affect reservoir water quality in tropical regions [2,3]. The prolonged dry season affects the water quality in terms of physical, chemical, and biological parameters [4,5]. In the rainy season, the reservoir water quality may deteriorate due to a combination of natural processes (rainfall intensity) and human activity [6]. In many cases, the deterioration of reservoir water quality is caused by complex pollution from both the watershed and the reservoir itself [7].

The Cirata Reservoir was built in 1988 for a 1008-megawatt (MW) hydropower plant, which supplied electricity for Java and Bali [8]. The other functions of the Cirata Reservoir are serving as a source of clean water and for irrigation, flood prevention, tourism, a site for aquaculture, and a conservation site for animals and plants [7]. The Cirata Reservoir is located in the middle of the Citarum Watershed [9] and is vulnerable to contamination by organic and inorganic pollutants from the Upstream Citarum Watershed [10].

For the last 30 years, the rapid economic growth in the Citarum Watershed spurred the conversion of green areas to open or built-up areas [11]. Deforestation, agricultural land clearing, urbanization, and construction lead to decreased water quality in the river [12]. Generally, the deterioration of water quality in the Citarum River is affected by agriculture, settlements, industry, livestock, and fisheries [13].

Agricultural activity is a main source of NO₃-N and phosphate (PO₄) flow to the Citarum Watershed [14]. Generally, farmers use urea (N) intensively, resulting in a very high nitrogen load in the Citarum Watershed [15]. Agricultural activity is also a main factor in soil erosion in the Upstream Citarum Watershed [16]. Land use conversions, such as forests into agricultural land and paddy fields, can increase the total suspended solids (TSS) in the Cirata Reservoir [17].

The high population in Bandung City contributes to a high nitrogen loading in the Citarum Watershed [15]. Domestic waste, such as feces and detergents, are the biggest sources of N and P [18]. The lack of an adequate sewage treatment system for domestic wastewater causes heavy flows of nitrogen, PO₄, and chemical oxygen demand (COD) into the Citarum River [19]. In addition, the increased settlement in the upper Citarum Watershed has had an effect on the dissolved oxygen (DO) decline [20].

It was estimated that 2700 industries were in the Citarum Watershed in 2018, which left 2800 tons of waste per day [21]. The disposal of textile industry effluent without a treatment process results in a high COD and biochemical oxygen demand (BOD) in the Citarum River [22]. This creates a dilemma for nearby inhabitants. For example, people in an industrial area in Majalaya compromise with poor water quality out of necessity because their income source relies on textile factories in their living area [23].

The cattle farm released 400 tons of cattle manure/day into the Citarum River [22]. Limited time and land are the farmers’ reasons for not processing the cattle manure into biogas, fertilizer, or media for growing worms because several farmers do not have large areas of land [24]. Because of that, the cattle manure is transported by runoff to the river, and the nutrient concentration in the Citarum River increases.

Previously, the number of floating cage aquacultures in 1988 in the Cirata Reservoir was only 74 cages [25], but recently, it has increased to 85,393 cages [8]. The promising economic profits encourage fishers to feed their fish excessively to accelerate growth [26]. However, the uneaten pellets precipitate to the bottom of the reservoir [7], increasing PO₄ and eutrophication in the Cirata Reservoir [27].

The purpose of this study was to identify the characteristics and changes in water quality during the rainy and dry seasons in the Cirata Reservoir. The differences in water quality, climate, and hydrology parameters were tested using the Mann–Whitney U-test. A principal component analysis (PCA) was used to determine the water quality characteristics.

2. Materials and Methods

This research started with sampling and analysis of the water quality, climate, and hydrology in the Cirata Reservoir during the rainy and dry seasons. Afterward, different tests were used to identify significantly different parameters between the rainy and dry seasons. Then, each parameter was processed using the PCA method to obtain the water quality characteristics in each season. A flowchart of the research methodology is presented in Figure 1.

2.1. Research Area

The research area was located in the Cirata Reservoir, which is a part of the Middle Citarum Watershed. Administratively, the research area location is located in the Cianjur Regency (districts of Cikalong Kulon, Mande, Sukaluyu, Ciranjang, and Haurwangi), Purwakarta Regency (districts of Citamiang and Maniis), and Bandung Barat Regency (Cipeundeuy District). Geographically, the research location is located at 06°41′02.70″ S–06°47′51″ S and 107°14′42.13″ E–107°22′04.77″ E. A map of the research location is presented in Figure 2.

2.2. Data Source

This research used primary data collected through sampling activity in the Cirata Reservoir. The 12 sampling sites were located in the Cirata Reservoir, where the samplings were carried out during the dry season (on 21 September 2019) and the rainy season (on 18 March 2020). The collected data consisted of water quality, climate, and hydrology parameters. The sampling sites are presented in Figure 3 and

Table 1. Information for sampling locations.

Site	Coordinate		Location	Sampling Location Characteristic
	Longitude	Latitude
D-1	107°16′13.1″ E	06°45′59″ S	Cisokan River Mouth	River Mouth
D-2	107°17′11.9″ E	06°46′31.7″ S	Citarum River Mouth	River Mouth
D-3	107°15′35.0″ E	06°44′35.5″ S	Cibalagung River Mouth	River Mouth
D-4	107°15′06.1″ E	06°43′09″ S	Cikundul River Mouth	River Mouth
D-5	107°17′06.2″ E	06°42′19.1″ S	Cimanggu River Mouth	River Mouth
D-6	107°20′30.3″ E	06°45′05.5″ S	Nyenang	River Mouth
D-7	107°17′17.2″ E	06°45′43.5″ S	Pasir Pogor	Middle Reservoir
D-8	107°18′41.8″ E	06°44′27.1″ S	Sangkali	Middle Reservoir
D-9	107°18′45.2″ E	06°43′51.6″ S	Rawa Buaya	Middle Reservoir
D-10	107°19′13″ E	06°43′16.5″ S	Plumbon	Middle Reservoir
D-11	107°17′27.1″ E	06°43′48.7″ S	Plumbon Mosque	Middle Reservoir
D-12	107°15′47″ E	06°43′34.9″ S	Ciputri	Middle Reservoir

. Some images of the sampling locations during the survey activities are presented in Figure 4.

2.3. Analysis

This research analyzed the water quality, climate, and hydrology of the Cirata Reservoir. The water quality parameters consisted of transparency, salinity, TSS, pH, water temperature, DO, ammonia (NH₃-N), NO₃-N, PO₄, and nitrate and phosphate in the sediment. The climate parameters consisted of air temperature, solar radiation, and wind velocity. The hydrological parameters consisted of water flow rate and wave height.

Each water quality sample was tested in the field and then analyzed in the laboratory. The climate parameter measurements used a heat index WGT meter for air temperature, a Lutron LX—101A light meter for solar radiation, and an anemometer for air velocity. The measurement of the water flow rate parameter was performed using a current meter. The wave height parameter was measured using a HOBO water-level logger.

2.4. Statistical Analysis

The differences between water quality, hydrology, and climate parameters in the Cirata Reservoir were analyzed using the Mann–Whitney U-test method [30]. The Mann–Whitney U-test refers to equality below [31]:

$$U=\min \left\{U_1,U_2\right\},$$

(1)

$$U_1=n_1{n}_2+\frac{n_1\left(n_1+1\right)}{2}-R_1,$$

(2)

$$U_2=n_1{n}_2+\frac{n_2\left(n_2+1\right)}{2}-R_2,$$

(3)

where:

• $U_1$ = The total number of sample I observations preceding sample II observations;
• $U_2$ = The total count of sample II observations preceding sample I observations;
• $n_1$ or $n_2$ = The size of sample I (X₁) or sample II (X₂);
• $R_1$ or $R_2$ = The rank sums of samples X₁ or X₂.

The water quality, climate, and hydrology data from each season were processed by PCA. The PCA results show the most useful parameters for explaining all data interpretations, reducing the data, and summarizing the statistical correlations between the parameters without reducing the original information [32]. The PCA equality is as follows [33]:

$$Z_{ij}=a_{i1}{X}_{1j}+a_{i2}{X}_{2j}+\cdots +a_{im}{X}_{mj}$$

(4)

where:

• Z = The component score;
• a = The component loading;
• X = The measured value of the variable;
• i = The component number;
• j = The sample number;
• m = The total number of variables.

3. Results

3.1. Water Quality

The variations in water temperature, transparency, salinity, TSS, pH, DO, NH₃-N, NO₃-N, nitrate and phosphate in the sediment during the dry season and rainy season in the Cirata Reservoir are presented in

Table 2. Analysis results of water quality, climate, and hydrology parameters in the Cirata Reservoir during the dry season.

Parameter	Unit	Site												Average
		D-1	D-2	D-3	D-4	D-5	D-6	D-7	D-8	D-9	D-10	D-11	D-12
A. Water Quality
Physical Parameters
Water Temperature	°C	31	30.5	30.4	32.5	32.5	30	29.9	29.7	30.6	29.4	30.6	29.6	30.55833
Transparency	cm	62	70	50	15	45	92	75	85	85	80	80	57	66.33333
Salinity	%	0.03	0.02	0.02	0.03	0.02	0.02	0.03	0.02	0.02	0.02	0.02	0.02	0.0225
TSS	mg/L	2.9	3.3	6.8	5.4	3.6	6.5	7.1	10	4.3	3.3	3.6	5.9	5.225
Chemical Parameters
pH	pH	7.28	7.48	7.15	7.41	8.22	7.24	7.65	7.24	7.96	7.45	8.21	7.59	7.573333
DO	mg/L	3.8	3.7	3.9	3.9	5.4	2	3.8	4.4	5.6	4	4.7	5.4	4.216667
NH3-N	mg/L	0.007	0.027	0.013	0.024	0.132	0.013	0.04	0.013	0.078	0.025	0.119	0.03	0.043417
NO3-N	mg/L	0.014	0.314	0.284	0.54	0.111	0.75	0.149	0.089	0.374	0.66	0.337	0.104	0.3105
Phosphate (PO4)	mg/L	0.081	0.055	0.052	0.052	0.119	0.047	0.047	0.047	0.036	0.039	0.06	0.039	0.056167
Sediment Content
Nitrate	mg/kg	4.503	5.311	5.282	5.664	4.752	5.117	6.092	5.443	5.726	6.21	5.668	5.904	5.472667
Phosphate	mg/kg	7.75	6.39	4.95	7.22	6.24	5.35	6.44	7.18	7.45	7.53	7.66	6.48	6.72
B. Climate
Air Temperature	°C	32.7	33.6	31.9	39.4	37.6	29	35	31	34.7	33	36	30.5	33.7
Solar Radiation	lux	91,200	9700	88,000	42,000	42,600	1130	92,800	31,100	38,000	63,000	69,300	42,000	50,902.5
Wind Velocity	m/s	1.105	1.26	0.49	2.75	2.52	0.98	1.61	6.7775	4.37	4.05	3.61	4.985	2.875625
C. Hydrology
Water Flow Rate	m/s	0.033	0.014286	0.05	0.5	0.02	0.009756	0.046875	0.05	0.1	0.06	0.046	0.037	0.080576
Wave Height	cm	0.5	1.4	0.3	0.3	0.5	0.5	0.6	4.9	3.8	1.5	4.6	1	1.658333

and

Table 3. Analysis results of water quality parameters in the Cirata Reservoir in the rainy season.

Parameter	Unit	Site												Average
		D-1	D-2	D-3	D-4	D-5	D-6	D-7	D-8	D-9	D-10	D-11	D-12
A. Water Quality
Physical Parameters
Water Temperature	°C	31	31	29	28	33	30	30	30	28	29	28	32	29.91667
Transparency	cm	85	85	65	30	55	95	80	75	70	90	80	80	74.16667
Salinity	%	0.01	0.01	0	0.02	0.01	0.01	0.01	0.02	0.02	0.02	0.01	0.01	0.0125
TSS	mg/L	6.8	7.3	55	80	29	35	12	8.4	4.6	4.6	18	10	22.55833
Chemical Parameters
pH	pH	7.92	8.81	7.92	8.03	8.56	8.1	8.2	7.84	8.44	7.55	8.83	9.05	8.270833
DO	mg/L	4.1	3.9	4.1	3.8	4.8	3.8	3.9	5.3	4.1	3	4.1	3.4	4.025
NH3-N	mg/L	0.0054	0.0234	0.0054	0.0048	0.0246	0.016	0.0176	0.0105	0.0415	0.0091	0.092	0.1848	0.036258
NO3-N	mg/L	1.051	0.577	1.051	1.276	0.938	0.818	1.156	1.456	1.449	1.464	1.637	1.449	1.1935
Phosphate (PO4)	mg/L	0.0571	0.0727	0.0571	0.1143	0.1844	0.1844	0.1974	0.1948	0.2364	0.3195	0.2857	0.2468	0.179217
Sediment Content
Nitrate	mg/kg	2.52	1.08	2.44	3.5	1.52	1.64	1.28	4.04	2.96	2.8	2.3	3.54	2.468333
Phosphate	mg/kg	10.74	7.16	10.4	9.62	8.12	6.3	5.78	17.16	11.6	10.04	7.46	6.38	9.23
B. Climate
Air Temperature	°C	33.4	32	40.1	31	33.1	30.6	31.9	31	28.3	28.4	28.3	37	32.09167
Solar Radiation	lux	1140	1289	934	1088	672	265	499	437	441	470	326	1024	715.4167
Wind Velocity	m/s	1.36	3.04	0.685	1.35	1.25	1.275	1.295	3.94	4.065	5.05	0.7	1.25	2.105
C. Hydrology
Water Flow Rate	m/s	0.071	0.1	0.036	0.2	0.027	0.055	0.0625	0.086	0.09	0.13	0.031	0.031	0.076625
Wave Height	cm	1.3	3.3	0.4	5.9	0.7	4.6	2.6	10.1	4.6	5.2	7.5	5.9	4.341667

, respectively. The water quality analysis showed that the average values of water temperature, salinity, DO, NH₃-N, and nitrate in the sediment were higher in the dry season. The average values of air temperature, salinity, DO, NH₃-N, and nitrate in the sediment in the dry season were 30.55 °C, 0.0225%, 4.21 mg/L, 0.043 mg/L, and 5.47 mg/kg, respectively, and in the rainy season, they were 29.91 °C, 0.0125%, 4.025 mg/L, 0.036 mg/L, and 2.46 mg/kg, respectively.

In contrast, transparency, pH, NO₃-N, PO₄, and phosphate in the sediment were higher during the rainy season. The average values of transparency, pH, NO₃-N, PO₄, and phosphate in the sediment in the rainy season were 74.16 cm, 8.27, 1.19 mg/L, 0.18 mg/L, and 9.23 mg/kg, respectively, and in the dry season, they were 66.33 cm, 7.57, 0.31 mg/L, 0.056 mg/L, and 6.72 mg/L, respectively.

The different test results showed that the parameters of salinity, TSS, pH, NO₃-N, PO₄, nitrate and phosphate in the sediment were significantly different (α < 0.05) between the dry and rainy seasons. The different test results for water quality parameters are summarized in

Table 4. Summary of the Mann–Whitney U-test results for water quality, climate, and hydrology parameters in the Cirata Reservoir.

Parameter	Unit	Average in Dry Season	Average in Rainy Season	α Significance (α < 0.05)
Water Quality
Water Temperature	°C	30.55	29.91	0.319
Transparency	cm	66.33	74.16	0.378
Salinity	%	0.0225	0.0125	0.001
TSS	mg/L	5.225	22.55	0.001
pH	pH	7.57	8.27	0.002
DO	mg/L	4.21	4.025	0.671
NH3-N	mg/L	0.043	0.036	0.219
NO3-N	mg/L	0.31	1.19	0.0001
PO4	mg/L	0.056	0.179	0.0001
Nitrate in Sediment	mg/kg	5.47	2.46	0.0001
Phosphate in Sediment	mg/kg	6.27	9.23	0.045
Climate
Air Temperature	°C	33.7	32.09	0.178
Solar Radiation	lux	50,902.5	715.41	0.0001
Wind Velocity	m/s	2.87	2.1	0.478
Hydrology
Water Flow Rate	m/s	0.08	0.076	0.198
Wave Height	cm	1.65	4.34	0.014

Note: The gray color indicates significant differences in parameters between the two seasons.

.

3.2. Climate Conditions

The climate parameters in this research consisted of air temperature, solar radiation, and wind velocity in the Cirata Reservoir. The climate parameter conditions at each sampling site during the dry season and rainy season are presented in Table 2 and Table 3, respectively. Generally, the average values of air temperature, solar radiation, and wind velocity in the dry season were higher than in the rainy season. The average values of air temperature, solar radiation, and wind velocity in the dry season were 33.70 °C, 50,902.5 lux, and 2.87 m/s, respectively, and in the rainy season, they were 32.09 °C, 715.41 lux, and 2.10 m/s, respectively.

The Mann–Whitney U-test showed that solar radiation was significantly different (α < 0.05) between the dry and rainy seasons. The different test results for the climate parameters are summarized in Table 4.

3.3. Hydrological Conditions

The hydrological parameters in this research consisted of water flow rate and wave height. The hydrological parameter conditions in each location in the dry season and rainy season are presented in Table 2 and Table 3, respectively. Generally, the average water flow rate was higher in the dry season: 0.08 m/s in the dry season and 0.07 m/s in the rainy season. In contrast, the average wave height was higher in the rainy season: 4.34 cm in the rainy season and 1.65 cm in the dry season.

Based on the different test results, the wave height was significantly different (α < 0.05) between the dry and rainy seasons. The different test results for the hydrological parameters are summarized in Table 4.

3.4. Water Quality Characteristics in the Dry Season and Other Affected Parameters

PCA was used to determine the water quality characteristics in the Cirata Reservoir. Six principal components (PCs) were obtained that contained eigenvalue > 1 and explained 91.668% of the total variance, as shown in

Table 5. Eigenvalues of PCA in the dry season at the Cirata Reservoir.

Component	Eigen Value
	Total	% Variance	Cumulative (%)
1	4.480	27.998	27.998
2	3.667	22.920	50.918
3	2.425	15.156	66.074
4	1.761	11.004	77.077
5	1.324	8.273	85.351
6	1.011	6.318	91.668
7	0.631	3.945	95.613
8	0.324	2.025	97.637
9	0.218	1.364	99.001
10	0.109	0.683	99.684
11	0.051	0.316	100.000

Note: The gray color shows components with eigenvalue > 1.

. In addition, the PCA results for the dry season are summarized in

Table 6. Principal component (PC) loadings for the water quality, climate, and hydrology parameters of the Cirata Reservoir during the dry season.

Parameter	Component
	1	2	3	4	5	6
Water Temperature	0.877
Transparency	−0.637
Salinity		−0.536
TSS	−0.540				−0.605
pH	0.641	0.627
DO		0.687
NH3-N	0.664	0.566
NO3-N				−0.818
PO4	0.727		−0.566
Nitrate in Sediment			0.648
Phosphate in Sediment			0.460			0.553
Air Temperature	0.873
Solar Radiation				0.704
Wind Velocity		0.778
Water Flow Rate			0.804
Wave Height		0.842

and Figure 5.

PC1 accounted for 27.998% of the total variance. PC1 showed significant positive loadings for water temperature, pH, NH₃-N, PO₄, and air temperature but significant negative loadings for transparency and TTS. The PC1 characteristics were affected by the dry season climate and floating cage aquaculture. The dry season climate was characterized by a high air temperature, high water temperature, and low rainfall. The low rainfall influenced the reduction of sedimentation and solid material transportation to the Cirata Reservoir, thus the TSS tended to be low. The uneaten pellets from floating cages increased the PO₄ in the reservoir water. The high PO₄ concentration stimulated algae growth that caused a greenish opaque color in the water, and thus, the water transparency tended to be low. Algal photosynthesis increases the pH in the reservoir water body. The incomplete degradation of uneaten pellets and dead fish promoted high NH₃-N levels in the reservoir water. Accordingly, the high NH₃-N concentration increased the pH in the reservoir water.

PC2 explained 22.920% of the total variance. PC2 showed significant positive loadings for pH, DO, NH₃-N, wind velocity, and wave height. In contrast, salinity had a significant negative loading. The dry season climate and floating cage aquaculture controlled the PC2 characteristics. The high wind velocity in the dry season increased the wave height in the Cirata Reservoir. The low rainfall caused low salinity due to the reduction in the inflow of potassium chloride (KCI) fertilizer residues from agricultural land and limestone dissolution from the Citatah Karst. The floating cage aquaculture caused uneaten pellet deposition and increased algae growth in the Cirata Reservoir. Algal photosynthesis increased the pH and DO in the reservoir water. The accumulation of uneaten pellets resulted in high NH₃-N levels in the reservoir water.

Approximately 15.156% of the total variance was described in PC3. A significant positive loading was shown in PC3 for the parameters of nitrate and phosphate in the sediment and water flow rate. In contrast, PO₄ had a significant negative loading. The PC3 characteristics were affected by the floating cage aquaculture and water movement. The uneaten pellets settled to the bottom of the floating cage, where they increased nitrate and phosphate in the sediment. The rapid water flow distributed the uneaten pellets around the reservoir bottom, which increased the nitrate and phosphate concentrations in the sediment. PO₄ tends to settle into reservoir sediments, and thus, its concentration decreases in the reservoir water.

PC4 described 11.004% of the total variance. PC4 showed a significant positive loading for solar radiation but a significant negative loading for NO₃-N. PC4 was controlled by the dry season climate. Sunny and cloudless weather resulted in high solar radiation around the Cirata Reservoir, and low rainfall in the dry season reduced NO₃-N input from the Citarum Watershed.

PC5 represented 8.273% of the total variance. In PC5, the TSS had a significant negative loading. The low TSS was affected by low rainfall during the dry season because rainfall is a carrier of sediment material to the Cirata Reservoir.

PC6 explained 6.318% of the total variance. PC6 showed a significant positive loading for phosphate in the sediment. The high phosphate levels in the sediment were due to the deposition of uneaten pellets from floating cage aquiculture during the dry season.

3.5. Water Quality Characteristics in the Rainy Season and Other Affected Parameters

In the PCA results, there were six PCs that had an eigenvalue > 1, which explained 91.879% of the total variance (

Table 7. Eigenvalues of PCs for the rainy season at the Cirata Reservoir.

Component	Eigen Value
	Total	% Variance	Cumulative (%)
1	5.036	31.477	31.477
2	3.279	20.493	51.970
3	2.163	13.517	65.486
4	1.789	11.181	76.668
5	1.386	8.665	85.333
6	1.047	6.546	91.879
7	0.550	3.436	95.315
8	0.406	2.535	97.850
9	0.220	1.374	99.224
10	0.084	0.526	99.750
11	0.040	0.250	100.000

Note: The gray color shows components with an eigenvalue > 1.

). The PCA results of the rainy season are summarized in

Table 8. Principal component (PC) loadings for the water quality, climate, and hydrology parameters of the Cirata Reservoir during the rainy season.

Parameter	Component
	1	2	3	4	5	6
Water Temperature	−0.567
Transparency		0.645	−0.561
Salinity	0.884
TSS		−0.670
pH		0.593				0.497
DO				0.813
NH3-N		0.731	0.633
NO3-N	0.717
PO4	0.610	0.678
Nitrate in Sediment	0.638		0.593
Phosphate in Sediment	0.590			0.645
Air Temperature	−0.704
Solar Radiation	−0.509				0.675
Wind Velocity	0.709
Water Flow Rate	0.507	−0.626
Wave Height	0.774

and Figure 6.

PC1 accounted for 31.477% of the total variance. PC1 showed significant positive loadings for salinity, NO₃-N, PO₄, nitrate and phosphate in the sediment, wind velocity, water flow rate, and wave height, but significant negative loadings for water temperature, air temperature, and solar radiation. These patterns showed that the rainy season climate and the land use of the Citarum Watershed controlled PC1. In the rainy season, the high rainfall and wind velocity promoted increased water flow velocities and wave heights in the Cirata Reservoir. Agriculture, settlements, livestock, and the Citatah Karst in the Upstream Citarum Watershed affected the water quality of the Cirata Reservoir. The high rainfall flowed fertilizer residue, domestic waste, and livestock manure into the Cirata Reservoir, increasing NO₃-N, PO₄, nitrate and phosphate in the sediment, as well as its salinity. The increased salinity in the Cirata Reservoir is related to the contact of runoff with the Citatah Karst. Cloudy weather conditions in the rainy season decreased solar radiation, air temperature, and water temperature.

PC2 described 20.493% of the total variance. PC2 had a significant positive loading for transparency, pH, NH₃-N, and PO₄. In contrast, TSS and water flow rate had significant negative loadings. This indicated that PC2 was related to runoff from the Citarum Watershed and increased water volumes during the rainy season. The runoff brought fertilizer residue, domestic waste, and livestock manure from the Citarum Watershed, which increased NH₃-N and PO₄ concentrations in the Cirata Reservoir. The runoff drove limestone dissolution in the Citatah Karst and transportation of calcium carbonate (CaCO₃), increasing the pH in the Cirata Reservoir. The increase in water volume from rainfall, runoff, and streamflow during the rainy season promoted dilution of the water body, which resulted in low TSS levels and high transparency.

PC3 represented 13.517% of the total variance and showed significant positive loadings for NH₃-N and nitrate in the sediment. PC3 was related to high rainfall and anthropogenic activities in the Citarum Watershed. Fertilizer residue, domestic waste, and livestock manure from the Citarum Watershed were pushed downstream by rainfall, increasing NH₃-N and nitrate in the sediment in the Cirata Reservoir.

PC4 explained 11.181% of the total variance. PC4 described significant positive loadings for DO and phosphate in the sediment. The high levels of DO and phosphate in the sediment were related to the land use in the Citarum Watershed and the rainy season climate. The high DO in the Cirata Reservoir was affected by the cold reservoir water and water inflow during the rainy season because oxygen in the atmosphere was more easily absorbed by cold water. Moreover, the high PO₄ concentration in the sediment was affected by fertilizer residue and domestic waste that were carried away by rainfall to the Cirata Reservoir.

PC5 explained 8.665% of the total variance and showed a significant positive loading for solar radiation. The high solar radiation showed that the samples collected at several locations were collected in sunny conditions.

PC6 described 6.546% of the total variance and had a significant positive loading for pH. The high pH in the rainy season was related to the Citatah Karst southeast of the Cirata Reservoir. The high runoff stimulated the limestone dissolution process in the Citatah Karst. Consequently, the CaCO₃ flowed to the Cirata Reservoir, which increased the pH.

4. Discussion

4.1. Water Quality Dynamics in the Dry Season

The water quality of the Cirata Reservoir in the dry season was controlled by the climate conditions and floating cage aquaculture. Solar radiation, air temperature, water temperature, wind velocity, water flow rate, wave height, pH, DO, NH₃-N, PO₄, and nitrate and phosphate in the sediment were high in the dry season. On the contrary, transparency, salinity, TSS, and NO₃-N were low.

The high solar radiation in the dry season increased the water temperature and air temperature in the Cirata Reservoir because solar radiation is the greatest source of heat for the reservoir [34]. The high wind velocity affected the increased water flow rate and wave height in the Cirata Reservoir. Furthermore, the low rainfall in the dry season influenced water quality in the Cirata Reservoir as a result of reduced runoff and material erosion from the Citarum Watershed.

The low TSS in the Cirata Reservoir during the dry season was related to the climate conditions, especially rainfall intensity. High rainfall distributes various solid materials into rivers and reservoirs, increasing the TSS [1,35]. The low rainfall in the dry season resulted in a low TSS level in the Cirata Reservoir because the rainfall acts as a medium for solid material transportation [36].

The low rainfall influenced the decrease in salinity in the Cirata Reservoir. The salinity concentration was related to the agricultural activity in the Upstream Citarum Watershed and the Citatah Karst located southeast of the Cirata Reservoir. Agricultural activities using KCI fertilizer could increase chloride (Cl⁻) and salinity in the soil [37]. The Citatah Karst is composed of limestone [29], and its composition is dominated by CaCO₃ [36]. The inflow of Cl⁻ from agricultural activities and CaCO₃ from the Citatah Karst was decreased during runoff reduction in the dry season. Therefore, the salinity concentration in the dry season tended to be low.

During the dry season, the low rainfall was one of the factors that reduced the NO₃-N concentration in the Cirata Reservoir. The land use of the Upstream Citarum Watershed is dominated by agriculture and settlements [16]. Agricultural activities affect the NO₃-N concentrations in water [12]. The high rainfall can transport any material from agricultural land in the upstream watershed, which increased the nutrient levels in the reservoir water [1,2,6]. Conversely, the low rainfall decreased the NO₃-N distribution to the Cirata Reservoir. This pattern similarly happened in Thailand, where the low rainfall in the dry season resulted in low NO₃-N levels in the Kwan Phayao Reservoir [2].

The high PO₄ concentration during the dry season in the Cirata Reservoir was influenced by floating cage aquaculture. Most of the fisher continuously fed the fish until they were full [26]. The uneaten pellets accumulate as organic materials, which increase the PO₄ concentrations in the Cirata Reservoir [27].

The high PO₄ changed the Cirata Reservoir’s trophic status to eutrophic [10], and it even reached a hypertrophic state in 2015 [27]. The eutrophic conditions drove algal blooms [4] because PO₄ is a key element for algal growth [1]. The rapid growth of algae increased organic turbidity [4], physically changing the water color to blue or greenish [38]. It caused low transparency in the Cirata Reservoir.

The high NH₃-N concentration in the dry season in the Cirata Reservoir was related to dense floating cages. The deposition of uneaten pellets disturbed aerobic and anaerobic bacteria activity, thus the incompletely degraded pellets produced NH₃-N, which is dangerous for aquatic organisms [39]. A large number of floating cages in the Cirata Reservoir also caused oxygen depletion in the water [8], and it potentially caused fish deaths. Fish death in the Cirata Reservoir occurs at a DO level under 2 mg/L, and it also increases the NH₃-N levels around the floating cages up to between 0.798 mg/L and 1.6 mg/L [40].

The increased pH and DO during the dry season in the Cirata Reservoir were influenced by the floating cage aquaculture. The deposition of uneaten pellets increased nutrient levels in the water, which drove algal growth [41]. During the photosynthesis process, algae consume carbon dioxide (CO₂), increasing the pH and DO in the water [42]. The photosynthesis processes of aquatic autotrophs are the main source of increasing oxygen levels in the water [43]. In addition, the accumulation of incompletely degraded pellets increases NH₃-N in the water [39]. Generally, NH₃-N is alkaline [44] and, therefore, increases the pH in the Cirata Reservoir.

The floating cage aquaculture and water current increased phosphate and nitrate levels in the sediment. The uneaten pellet deposits increase the sedimentation of organic material containing nitrogen and PO₄ onto the reservoir floor [7,39,41]. The water current can push sediment resuspension to quickly redeposit it around the reservoir floor [33], and the uneaten pellet deposits also spread to almost the entire reservoir bottom. Consequently, NO₃-N and PO₄ concentrations are increased in the reservoir sediments.

During the dry season, the conditions of the Cirata Reservoir area played an important role in the reservoir water quality. High solar radiation and low rainfall affected the water quality of the Cirata Reservoir. In addition, the floating cage aquaculture increased nutrient levels, pH, and DO and also decreased transparency.

4.2. Water Quality Dynamics in the Rainy Season

The climate conditions and water flow of the Citarum Watershed affected the water quality of the Cirata Reservoir in the rainy season. During this season, the transparency, salinity, pH, DO, NH₃-N, NO₃-N, PO₄, nitrate and phosphate in the sediment, wind velocity, water flow rate, and wave height were high. On the contrary, TSS, solar radiation, air temperature, and water temperature were low in the rainy season.

The cloudy weather during the rainy season decreased solar radiation, air temperature, and water temperature. The high rainfall and wind velocity increased the water flow rate and wave height in the Cirata Reservoir. The high DO was also influenced by the rainy season. The low air and water temperatures during the rainy season increased DO in the Cirata Reservoir. Generally, oxygen from the atmosphere is more easily absorbed by low-temperature water [34,44]. In addition, the high DO in the reservoir was also caused by the inflow of cold oxygenated waters from the watershed in the rainy season [45].

The high rainfall became a medium for waste transportation from the Citarum Watershed to the Cirata Reservoir [38]. Agriculture and settlements dominated the land use of the Citarum Watershed [16]. Generally, farmers in the Upstream Citarum Watershed use inorganic fertilizers that contain nitrogen and PO₄ [10]. Moreover, settlements along the Citarum River did not have adequate sanitation facilities [22], resulting in feces, urine, and detergents, which contain nitrogen (N) and phosphorus (P), to flow into the river [18]. The nutrient input was aggravated by livestock activities because about 57% of farmers in the Citarum Watershed dump their livestock manure directly without processing it [24]. The high rainfall transported agriculture, domestic, and livestock waste to the Cirata Reservoir, increasing NH₃-N, NO₃-N, PO₄, and nitrate and phosphate in the sediment in the Cirata Reservoir. This is similar to the patterns in the Kwan Phayao Reservoir, Thailand; the Adolfo López Mateos Reservoir, Mexico; and the Irapé Hydroelectric Power Plant Reservoir, Brazil. These areas also showed high nitrogen and PO₄ concentrations, which were related to agricultural runoff during the rainy season [1,2,6].

The high salinity during the rainy season was related to the agricultural activity in the Citarum Watershed and the Citatah Karst southeast of the Cirata Reservoir. The application of KCl fertilizer increased Cl⁻ and salinity in the soil [37]. The Citatah Karst is composed of limestone [29], and its composition is dominated by CaCO₃ [36]. The reaction of water with CaCO₃ resulted in calcium bicarbonate (Ca(HCO3)₂) [46]. Because of this, the confluence of runoff and limestone increased the salinity of the Citatah Karst. During the rainy season, the high runoff transported Cl⁻ from KCl fertilizer residues and Ca(HCO3)₂ from the Citatah Karst to the Cirata Reservoir, increasing the salinity. This is similar to the patterns in the Tombolo Dam, Tanzania, where the high salinity was affected by the limestone dissolution process and runoff from the upstream watershed in the rainy season [47].

The existence of the Citatah Karst was one of the factors for the high pH during the rainy season. The weathering of rocks that contain CaCO₃ drives carbonate transportation and pH increases in water bodies [48]. The high rainfall drove the Citatah Karst limestone weathering process and CaCO₃ transportation to the Cirata Reservoir, and increased its pH during the rainy season. This is similar to the situation in other tropical areas. For example, the pH was generally higher during the rainy season in a karst area in Yucatán, Mexico [49].

The high transparency and low TSS levels were affected by the water volume increase in the Cirata Reservoir during the rainy season. The high rainfall, runoff water, and streamflow diluted the water bodies, resulting in low TSS levels and high transparency. This is similar to the case in Bakun Reservoir, Malaysia [3]. The decrease in TSS concentrations and turbidity caused high transparency in the reservoir water. However, this is different from what occurred in the Atlanga Reservoir and Adolfo López Mateos Reservoir in Mexico. The high rainfall brought materials to the water body, increasing the TSS levels [1,35].

Land use in the Citarum Watershed and high rainfall were the main factors for water quality changes in the Cirata Reservoir during the rainy season. The inflow of water, fertilizer waste, livestock manure, domestic waste, and dissolved limestone from the Citarum Watershed to the Cirata Reservoir increased DO, nutrients, pH, and salinity and also decreased TSS levels in the reservoir.

The differences in water quality, climate, and hydrology between the rainy and dry seasons are summarized in

Table 9. Summary of analysis results for water quality, climate, and hydrology parameters in the rainy and dry seasons.

Parameter	Season	Unit	Average	Significant Difference (α < 0.05)	Loading Contribution for PC
A. Water Quality
Physical Parameters
Water Temperature	D	°C	30.55833	No	Positive
	R		29.91667		Negative
Transparency	D	cm	66.33333	No	Negative
	R		74.16667		Positive
Salinity	D	%	0.0225	Yes	Negative
	R		0.0125		Positive
TSS	D	mg/L	5.225	Yes	Negative
	R		22.55833		Negative
Chemical Parameters
pH	D	pH	7.573333	Yes	Positive
	R		8.270833		Positive
DO	D	mg/L	4.216667	No	Positive
	R		4.025		Positive
NH3-N	D	mg/L	0.043417	No	Positive
	R		0.036258		Positive
NO3-N	D	mg/L	0.3105	Yes	Negative
	R		1.1935		Positive
PO4	D	mg/L	0.056167	Yes	Positive
	R		0.179217		Positive
Sediment Content
Nitrate	D	mg/kg	5.472667	Yes	Positive
	R		2.468333		Positive
Phosphate	D	mg/kg	6.72	Yes	Positive
	R		9.23		Positive
B. Climate
Air Temperature	D	°C	33.7	No	Positive
	R		32.09167		Negative
Solar Radiation	D	lux	50,902.5	Yes	Positive
	R		715.4167		Negative
Wind Velocity	D	m/s	2.875625	No	Positive
	R		2.105		Positive
C. Hydrology
Water Flow Rate	D	m/s	0.080576	No	Positive
	R		0.076625		Positive
Wave Height	D	cm	1.658333	Yes	Positive
	R		4.341667		Positive

Note: D is the rainy season and R is the dry season.

. The table also illustrates some of the significantly different parameters and the loading contribution of each parameter.

5. Conclusions

The water quality characteristics in the Cirata Reservoir during the dry season were affected by the environmental conditions in the reservoir. The high solar radiation stimulated an increase in water temperature. Furthermore, the low rainfall resulted in low salinity and concentrations of TSS and NO₃-N in the Cirata Reservoir. The deposition of pellets from floating cages directly or indirectly increased pH, DO, NH₃-N, PO₄, nitrate and phosphate in the sediments and decreased transparency.

The water quality characteristics in the Cirata Reservoir during the rainy season were highly related to the environmental dynamics outside the reservoir, especially the Citarum Watershed conditions. The high rainfall transported agriculture, livestock, and domestic waste from the Citarum Watershed to the Cirata Reservoir, increasing salinity, pH, NH₃-N, NO₃-N, PO₄, and nitrate and phosphate levels in the sediment. Furthermore, the increased water inflow from the Citarum Watershed resulted in a high DO, high transparency, and low TSS levels in the Cirata Reservoir.

The water quality conditions in the dry season and rainy season showed that the floating cage aquaculture and land use of the Citarum Watershed had negative effects on the water quality of the Cirata Reservoir. The fishers using floating cages should be involved in education and training on efficient feeding methods and floating cage limitations because dense floating cages and overfeeding decrease fishery productivity [8].

The complex activity in the Citarum Watershed area should be managed using an environmentally and economically advantageous approach. Farmers should be involved in education on manure processing. Furthermore, farmers should be encouraged to use livestock manure for fertilizer and should be educated about efficient fertilizer application. In addition, any stakeholder should plant commercial trees along the Citarum levee that can serve as sediment traps and increase economic value.

Therefore, integrated management between the Citarum Watershed and the Cirata Reservoir is needed to maintain environmental sustainability in the Cirata Reservoir to continuously provide benefits for hydropower plants, fishers’ livelihoods, tourism, and a clean water supply.