Water quality parameters
The following water quality parameters were measured in the lake: conductivity, turbidity, chlorophyll-a (Chl-a), depth, DO, temperature, TP, PO43–, NO3, TN, SiO44–, NH4+. The data were collected monthly from the 9 stations from March 2008 to June 2019 and again from January 2020 to April 2021. Temperature, dissolved oxygen, and conductivity were measured in situ using a multiparametric field probe (YSI Incor.550, USA); turbidity with a turbidimeter (HACH 21000, Germany). The water depth at each sampling station was determined by an echo sounder (Plastimo Echotest II, 59588, France). Water samples for nutrients (TP, PO43–, NO3, TN, SiO44–, and NH4+) and Chl-a analyses were taken at each station using a small 5-liter polyethylene canister, previously rinsed with distilled water. Before the nutrient analyses, water samples were filtered using Macherey-Nägel GF/5 filters with a porosity of 0.7 μm (Bartram and Ballance 1996). Both filtered water and labeled vials were placed in a portable freezer at 4°C and later stored in a fridge at -20 °C for further nutrient and chlorophyll-a analyses in the laboratory. Phosphate, Ammonium, Nitrites, Silica, and Chl-a were analyzed on filtered water using various standardized techniques of UV-Visible spectrophotometric analysis of water samples (APHA, 2005; Rodier, 2009). The laboratory details of sample analysis procedures are described in Walumona et al . (2021).
In order to evaluate how lake level changes may affect the water quality of the Lake, a water quality index (WQI) was derived (Bhateria and Jain, 2016). The WQI included the parameters given in Table I and was calculated as an indicator of water pollution in two steps using the Bhateria and Jain (2016) method.
1°) determination of a subjective water quality index by using k coefficient:
\(\text{WQI}_{\text{sub}}=k\frac{\sum_{i=0}^{n}{\text{Ci}\ \text{Pi}}}{\sum_{i=0}^{n}\text{Pi}}\)(1)
Where, k is a subjective coefficient with a value ranging between 0.25 (mostly for heavily contaminated water indicated by blackish color, harsh odor, visible fermentation, etc.) and 1 (apparent contamination, clear or with natural suspended solids), n is the total number of parameters, Ci is the value assigned to parameter i after normalization and Pi is the relative weight assigned to each parameter with a value between 1 and 4, where 4 is assigned to a parameter that is most critical for the preservation of aquatic life (e.g. dissolved oxygen) and 1 is assigned to the parameter that has a less direct impact (e.g. temperature and pH) (Table I). Ci is a normalization factor, to make the WQI more objective and realistic (Herna’ndez-Romero et al ., 2004).
2°) Following normalization, an objective WQI was calculated using k = 1 in order to account only for variations due to the parameters measured in situ as:
\(\text{WQI}_{\text{obj}}=\frac{\sum_{i=0}^{n}{\text{Ci}\ \text{Pi}}}{\sum_{i=0}^{n}\text{Pi}}\)(2)