Predicting anoxia in the wet and dry periods of tropical semiarid reservoirs.

The dissolved oxygen (DO) level in the hypolimnion of lakes and reservoirs can reach anoxic conditions, which favor the release of phosphorus from the sediment bed to the water column. However, to estimate nutrient release from sediment is extremely important to quantify the duration of anoxia. In low latitude regions, the water-sediment layer is warmer than in temperate regions and eutrophication is usually more severe, potentially accelerating oxygen depletion and extending the anoxia period. Considering that the available equations to quantify the duration of anoxia were developed for temperate lakes, there is a need to effectively quantify this period in lakes and reservoirs located in other climate regions, such as the semiarid. In this study, the dynamics of thermal stratification was analyzed as a function of the Relative Water Column Stability coefficient (RWCS) and then correlated with DO dynamics for nineteen tropical semiarid reservoirs. RWCS values were higher during the rainy season, when anoxia duration was longer and the hypolimnion was thicker with respect to total water depth. Then, two new equations for quantification of anoxia duration, based on the equation originally developed for temperate climate, were adapted for the wet and dry seasons of the tropical semiarid region. The results showed that the proposed equations presented a better performance compared to the original one, which tends to underestimate anoxia in tropical semiarid reservoirs. This work intended to provide simple and locally adjusted tools to better quantify anoxic events and support the water quality and internal phosphorus load modeling for tropical semiarid reservoirs.

Assessment of phosphorus loading dynamics in a tropical reservoir with high seasonal water level changes.

Nutrient accumulation in man-made reservoirs has been documented worldwide. Therefore, quantifying phosphorus loading and understanding its temporal dynamics in reservoirs is mandatory for sustainable water management. In this study, the Vollenweider's complete-mix phosphorus budget model was adapted to account for high water level variations, which are a common feature in tropical reservoirs, and for internal phosphorus loads. First- and zero-order kinetics were adopted to simulate phosphorus settling and release from the sediment layer, respectively, considering variable area of phosphorus release according to the height of the anoxic layer. The modeling approach was applied for a 52-months period to a 31-years-old reservoir located in the semiarid region of Brazil with 7.7 hm3 storage capacity. The simulations were supported by hydrological, meteorological and water quality data, as well as analyses of phosphorus partitioning of the reservoir bed sediment. The external phosphorus load was estimated from a relationship adjusted between inflow and phosphorus concentration, revealing an u-shaped pattern. Sedimentary phosphorus linked to iron and aluminum (PFeAl) increased over time and along the reservoir. Such measurements were used to estimate the internal phosphorus load, i.e., the yield from the bed sediments to the water column. The adaptations proposed to the model's structure improved its capacity to simulate phosphorus concentration in the water column, from "not satisfactory" to "good". We estimate that the internal phosphorus load currently accounts for 44% of the total load. It prevailed during the wet period, when reservoir stratification and hypolimnetic hypoxia were more notable, resulting in higher phosphorus concentration in the water column due to the combined effects of internal and external loadings. However, if the reservoir were 70 years older, the internal load would reach 83% of the total and the reservoir would become a source instead of a sink of phosphorus.

Pond bank access as an approach for managing toxic cyanobacteria in beef cattle pasture drinking water ponds.

Forty-one livestock drinking water ponds in Alabama beef cattle pastures during were surveyed during the late summer to generally understand water quality patterns in these important water resources. Since livestock drinking water ponds are prone to excess nutrients that typically lead to eutrophication, which can promote blooms of toxigenic phytoplankton such as cyanobacteria, we also assessed the threat of exposure to the hepatotoxin, microcystin. Eighty percent of the ponds studied contained measurable microcystin, while three of these ponds had concentrations above human drinking water thresholds set by the US Environmental Protection Agency (i.e., 0.3 µg/L). Water quality patterns in the livestock drinking water ponds contrasted sharply with patterns typically observed for temperate freshwater lakes and reservoirs. Namely, we found several non-linear relationships between phytoplankton abundance (measured as chlorophyll) and nutrients or total suspended solids. Livestock had direct access to all the study ponds. Consequently, the proportion of inorganic suspended solids (e.g., sediment) increased with higher concentrations of total suspended solids, which underlies these patterns. Unimodal relationships were also observed between microcystin and phytoplankton abundance or nutrients. Euglenoids were abundant in the four ponds with chlorophyll concentrations > 250 µg/L (and dominated three of these ponds), which could explain why ponds with high chlorophyll concentrations would have low microcystin concentrations. Based on observations made during sampling events and available water quality data, livestock-mediated bioturbation is causing elevated total suspended solids that lead to reduced phytoplankton abundance and microcystin despite high concentrations of nutrients, such as phosphorus and nitrogen. Thus, livestock could be used to manage algal blooms, including toxic secondary metabolites, in their drinking water ponds by allowing them to walk in the ponds to increase turbidity.