Long-term monitoring of water quality in the monomictic Ishitegawa Reservoir indicates carbon limitation

Abstract

The monomictic Ishitegawa Reservoir (surface area: 0.50 km²; total capacity: 1.28 × 10⁷ m³; mean retention time: 0.33 y) is located on the Ishite River (N 33ﾟ53', E 132ﾟ50') in Ehime Prefecture, Japan. In this reservoir, summer chlorophyll a concentrations (Chl-a) have been observed to decrease gradually from year to year, but the cause remains unclear. To determine the cause, we first examined a monthly data set from January 1983 to January 2003 that contained measures of water temperature (WT), pH, Chl-a, total phosphorus and total nitrogen taken at 0.5 m water depth at the main station in the reservoir (Sta. R). We found two significant long-term linear regressions of pH against date: a positive relationship when pH≤ 8.10 (R² = 0.099, p < 0.001, n = 123) and a negative relationship when pH > 8.10 (R² = 0.164, p < 0.001, n = 118). This pattern was also found at a station near the head of the reservoir where river inflow occurs (Sta. H). Low pH (≤ 8.10) was observed mainly in the circulation period (October – March) and high pH (> 8.10) was found during times of stratification (April – September). Next, we examined the relationship of pH = 8.1 with other limnological parameters by using a monthly data set of dissolved inorganic carbon (DIC), dissolved oxygen, and main cation and anion concentrations from February 1993 to January 2003. Concentration of free CO₂ (dissolved CO₂ + H₂CO₃) was calculated from DIC, pH, WT and ionic strength by using the dissociation equations of carbonic acid. pH was negatively related to the log partial pressure of gaseous CO₂ in equilibrium with free CO₂ (R² = 0.968, p < 0.001, n = 120). At pH 8.1, free CO₂ was in equilibrium with atmospheric CO₂ (360 μatm partial pressure). Chl-a concentrations during the stratification period (pH > 8.10) decreased from year to year following the gradual decrease in the free CO₂ concentration and the gradual increase in pH in the preceding circulation period. This data suggests that partial carbon limitation of phytoplankton growth may have occurred under high pH (> 8.10). In addition, plastic-coated sheet fences that were installed across the upper 5 m of the water column near Sta. H during several intervals in the study period formed a strong interflow of river water below the fences, and may have strengthened carbon limitation at Sta. R.