Riparian Soil Pollution Caused by Sediment Metal Transport: Seasonal Changes and Ecological Risk Assessment.

The accumulation of pollutants in the sediment along surface water may negatively affect riparian zones and increase ecological risk. This article investigates the effects of metal sediments on riparian soil via field monitoring and ICP-OES analysis. To this end, pollution levels, seasonal changes, and potential sources of the pollutants were determined for the Melen River watershed, Turkey. The ecological statuses (contamination factor, enrichment factor, index of geo-accumulation, pollution index, modified pollution index, and potential and modified ecological risk indexes) of the watershed were also analyzed. Although no significant seasonal differences in the metal sediments were observed, their spatial distribution in the sediments and riparian soils varied markedly. Cr (11.4 to 136), Co (7.7 to 21.52), Cu (11.4 to 76.6), and Ni (14.06 to 128.2) recorded as mg/kg significantly increased from the upstream to the downstream. The metals possessing the highest risk in the sediment and riparian soil regarding the river health were Cu, Co, and Ni. The risk values were found to be heavily polluted (PI > 3 and MPI > 10), and the risk indexes were above the "desired environment without the risk". The risk index was found to be more than 50, and the modified risk indexes exceeded 200 at many points. The transportation of pollutants in surface water became evident in the sediment, resulting in adverse effects on the riparian zone and the ecological system.

Human health risk assessment of dissolved metals in groundwater and surface waters in the Melen watershed, Turkey.

Determination of metal risk levels in potable water and their effects on human health are vital in assessment of water resources. Risk assessment of metals to human health in a watershed, which has not been studied before, is the main objective of the present study. Surface and groundwater sampling was carried out between September 2010 and August 2011 in the Melen Watershed, Turkey, an important drinking water resource for millions of people. Metals were analyzed in the laboratory using inductively coupled plasma. Of the 26 different metals monitored, Al, B, Ba, Cr, Cu, Fe, Mn, Mo and V were found in surface water and As, B, Ba, Cr, Cu, Mn, Mo, V and Zn in groundwater. In groundwater, unitless hazard quotient (HQ) values were 6 for As, 2.7 for Mn and 1 for Zn, while in surface water all metals were below the risk level (HQ < 1). The ingestion risk was found to be higher than the dermal uptake risk. Arsenic mean concentration was observed to be 0.044 mg/L in groundwater. The As carcinogenic risk (CR) value was higher than the internationally accepted risk level (10(-4)) and with maximum ingestion of groundwater the carcinogenic risk was found to be higher in adults than children. These results show that even unpolluted watersheds can pose a risk to human health and that potential carcinogenic impacts should receive more attention.

Modeling biological oxygen demand of the Melen River in Turkey using an artificial neural network technique.

Artificial neural networks (ANNs) are being used increasingly to predict and forecast water resources' variables. The feed-forward neural network modeling technique is the most widely used ANN type in water resources applications. The main purpose of the study is to investigate the abilities of an artificial neural networks' (ANNs) model to improve the accuracy of the biological oxygen demand (BOD) estimation. Many of the water quality variables (chemical oxygen demand, temperature, dissolved oxygen, water flow, chlorophyll a and nutrients, ammonia, nitrite, nitrate) that affect biological oxygen demand concentrations were collected at 11 sampling sites in the Melen River Basin during 2001-2002. To develop an ANN model for estimating BOD, the available data set was partitioned into a training set and a test set according to station. In order to reach an optimum amount of hidden layer nodes, nodes 2, 3, 5, 10 were tested. Within this range, the ANN architecture having 8 inputs and 1 hidden layer with 3 nodes gives the best choice. Comparison of results reveals that the ANN model gives reasonable estimates for the BOD prediction.