Nutrient reference concentrations and trophic state boundaries in subtropical reservoirs.

Nutrient criteria as reference concentrations and trophic state boundaries are necessary for water management worldwide because anthropogenic eutrophication is a threat to the water uses. We compiled data on total phosphorus (TP), nitrogen (TN) and chlorophyll a (Chl a) from 17 subtropical reservoirs monitored from 2005-2009 in the São Paulo State (Brazil) to calculate reference concentrations through the trisection method (United States Environmental Protection Agency). By dividing our dataset into thirds we presented trophic state boundaries and frequency curves for the nutrient levels in water bodies with different enrichment conditions. TP and TN baseline concentrations (0.010 mg/L and 0.350 mg/L, respectively) were bracketed by ranges for temperate reservoirs available in the literature. We propose trophic state boundaries (upper limits for the oligotrophic category: 0.010 mg TP/L, 0.460 mg TN/L and 1.7 µg Chl a/L; for the mesotrophic: 0.030 mg TP/L, 0.820 mg TN/L and 9.0 µg Chl a/L). Through an example with a different dataset (from the Itupararanga Reservoir, Brazil), we encouraged the use of frequency curves to compare data from individual monitoring efforts with the expected concentrations in oligotrophic, mesotrophic and eutrophic regional systems. Such analysis might help designing recovery programs to reach targeted concentrations and mitigate the undesirable eutrophication symptoms in subtropical freshwaters.

Trophic state evolution in a subtropical reservoir over 34 years in response to different management procedures.

Despite their importance for water management, long-term studies on trophic state are relatively scarce in subtropical reservoirs. We analyzed total phosphorus (TP), total nitrogen (TN) and chlorophyll a (Chl a) concentrations in the Billings Reservoir (Brazil) over time: Phase 1 (1977-1992, Tietê River water was pumped to Billings to increase energy generation, 100 m(3) s(-1)); Phase 2 (1992-2007, Tietê water was conveyed to Billings only in special cases for flood avoidance, 8 m(3) s(-1)); and Phase 3 (2007-2010, besides flood control, Billings received Tietê water treated by an in situ flotation system, 13 m(3) s(-1)). We compared our results with data from 12 reservoirs to evaluate current (2005-2009) enrichment conditions. Phosphorus and nitrogen concentrations decreased (p < 0.05, MANOVA) from Phase 1 to 2 and were stable thereafter. TN/TP ratios increment (1977-2010) indicated shift from N- to P-limitation in the reservoir, affecting the phytoplankton. Nutrient levels in Billings are currently between the expected concentrations in mesotrophic and eutrophic reservoirs (0.03 mg L(-1) < TP < 0.42 mg L(-1), 0.8 mg L(-1) < TN < 7.6 mg L(-1)) and Chl a concentrations exceeded 34 µg L(-1), median for the eutrophic waterbodies from the dataset. Although water quality in Billings has improved, nutrient inputs from Tietê River pumping episodes, diffuse and internal sources are still favoring biomass accrual and compromising water uses.

On site flotation for recovering polluted aquatic systems: is it a feasible solution for a Brazilian urban river?

Pinheiros River (Brazil) plays a pivotal role in supplying water to Billings Reservoir, which presents multiple uses (human drinking, energy generation, irrigation, navigation, fishing and leisure). An intense monitoring program was performed during the years 2007 and 2008 to find out whether on site flotation is a feasible solution or not for improving the water quality of this urban river, attenuating the pollutants load caused by the water pumping to the reservoir (approximately 10 m³ s⁻¹). The monitoring of 18 variables (13,429 laboratorial analysis during the period of 490 days), suggested that despite the convenience of the on site approach for water treatment, especially for rivers located in fully urbanized areas, the flotation system is not enough itself to recover Pinheiros River water quality, given the several constraints that apply. Total phosphorus removal was high in percentage terms (about 90%), although the remaining concentrations were not so low (mean of 0.05 mg L⁻¹). The removal efficiency of some variables was insufficient, leading to high final mean concentrations of metals [e.g. aluminium (0.29 mg L⁻¹), chromium (0.02 mg L⁻¹) and iron (1.1 mg L⁻¹)] as well as nitrogen-ammonia (25.8 mg L⁻¹) and total suspended solids (18 mg L⁻¹) in the treated water.