Water Level Fluctuation Patterns
The lake water levels fluctuated by yearly amplitude (WLamp ± SD) of 2.25 ± 2.00 m from 1956 to 2020. Consequently, a LOWESS plot showed that the lake water level annual amplitude fluctuated during the period 1956 to 1975 and then decreased steadily up to its lowest level in 1989 ( amplitude = 0.1 m), with a subsequent increase to peak amplitude in 2008 (9.4, m) before a decreasing fluctuation between 2008 and 2021 (Figure 3). Linear regression analysis showed a non-significant (p = 0.12) negative relationship between yearly amplitude and time while the waveform Sine 3 parameter modeled as Wlamp =2.328*sin (2*ℼ*year/43.45+6.28) unlikely revealed a significant result () even though p < 0.05 but r² = -0.64 indicating a poor fit than a horizontal line. The trends in water level fluctuations as measured by DLTM (Figure 4) were poorly explained by linear regression model (p = 0.1229, r2 = 0.059). However, the waveform Sine (3 parameter) model: DLTM = 1.376*sin (2*ℼ*year/(42.98 + 6.28)) was highly significant although with a weak fit (P < 0.0001, r² = 0.21), and indicated peak rise in water levels after every ⁓20 years (Figure 4).
The depth frequency plot (Figure 5a) showed that for 25% of the years (1956-2021), the average depth of the lake was about 2 m while, for 20% of the years, the average depth was 3 m. For less than 10% of the years, the mean depth of the lake reached 4 m. Additionally only for less than 5% of the years, the lake means depth ranged between 10-14 m during the period from 1956-2021. The lake depth-area relationship showed significant dependence (R2 = 0.74) of the lake depth on its area (A, km2) modeled as: depth = 0.179e0.019A; Figure 5b. Thus, in current conditions, a 1m increase in lake depth leads to a ~90 km2 increase in lake area with likely influence on the riparian communities through overflows.