Deep learning models to predict flood events in fast-flowing watersheds.

This study aims to explore the reliability of flood warning forecasts based on deep learning models, in particular Long-Short Term Memory (LSTM) architecture. We also wish to verify the applicability of flood event predictions for a river with flood events lasting only a few hours, with the aid of hydrometric control stations. This methodology allows for the creation of a system able to identify flood events with acceptable errors within several hours' notice. In terms of errors, the results obtained in this study can be compared to those obtained by using physics-based models for the same study area. These kinds of models use few types of data, unlike physical models that require the estimation of several parameters. However, the deep learning models are data-driven and for this reason they can influence the results obtained. Therefore, we tested the stability of the models by simulating the missing or wrong input data of the model, and this allowed us to achieve excellent results. Indeed, the models were stable even if several data were missing. This method makes it possible to lay the foundations for the future application of these techniques when there is an absence of geological-hydrogeological information preventing physical modeling of the run-off process or in cases of relatively small basins, where the complex system and the unsatisfactory modeling of the phenomenon do not allow a correct application of physical-based models. The forecast of flood events is fundamental for correct and adequate territory management, in particular when significant climatic changes occur. The study area is that of the Arno River (in Tuscany, Italy), which crosses some of the most important cities of central Italy, in terms of population, cultural heritage, and socio-economic activities.

Heavy Metal Contamination in an Industrially Affected River Catchment Basin: Assessment, Effects, and Mitigation.

The concentrations of some heavy metals (Fe, Zn, Mn, Cu, Mo, Pb, Cd) were measured in river waters, macrozoobenthos, and fish (Kura scrapers) from one of the most developed mining areas in Armenia, the Debed River catchment basin. In order to assess heavy metal contamination and its hydro-ecological and health effects, the macrozoobenthos quantitative and qualitative parameters, geo-accumulation index, and hazard index were determined. Microalgal extraction experiments were conducted to assess the microalgal remediation efficiency for heavy metal removal from mining wastewaters. The results showed that the rivers in many sites were polluted with different heavy metals induced by mining activities, which adversely affected macrozoobenthos growth and caused human health risks in the case of waters used for drinking purposes. However, the river fish, particularly Kura scrapers, were determined to be safe for consumption by the local people, as per the conditions of the evaluated fish ingestion rate. The results have shown that microalgal remediation, particularly with Desmodesmus abundans M3456, can be used for the efficient removal ~(62-100%) of certain emerging contaminants (Mn, Pb, Cu, Zn, Cd) from mining wastewater discharged in the Debed catchment basin.

Temporal trends for inflow of perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) to Tokyo Bay, Japan, estimated by a receptor-oriented approach.

We estimated inflow rates of perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) to Tokyo Bay, Japan, between February 2004 and February 2011 by a receptor-oriented approach based on quarterly samplings of the bay water. Temporal trends in these inflow rates are an important basis for evaluating changes in PFOS and PFOA emissions in the Tokyo Bay catchment basin. A mixing model estimated the average concentrations of these compounds in the freshwater inflow to the bay, which were then multiplied by estimated freshwater inflow rates to obtain the inflow rates of these compounds. The receptor-oriented approach enabled us to comprehensively cover inflow to the bay, including inflow via direct discharge to the bay. On a logarithmic basis, the rate of inflow for PFOS decreased gradually, particularly after 2006, whereas that for PFOA exhibited a marked stepwise decrease from 2006 to 2007. The rate of inflow for PFOS decreased from 730kg/y during 2004-2006 to 160kg/y in 2010, whereas that for PFOA decreased from 2000kg/y during 2004-2006 to 290kg/y in 2010. These reductions probably reflected reductions in the use and emission of these compounds and their precursors in the Tokyo Bay catchment basin. Our estimated per-person inflow rates (i.e., inflow rates divided by the estimated population in the basin) for PFOS were generally comparable to previously reported per-person waterborne emission rates in Japan and other countries, whereas those for PFOA were generally higher than previously reported per-person waterborne emission rates. A comparison with previous estimates of household emission rates of these compounds suggested that our inflow estimates included a considerable contribution from point industrial sources.

3D Clumped Cell Segmentation Using Curvature Based Seeded Watershed.

Image segmentation is an important process that separates objects from the background and also from each other. Applied to cells, the results can be used for cell counting which is very important in medical diagnosis and treatment, and biological research that is often used by scientists and medical practitioners. Segmenting 3D confocal microscopy images containing cells of different shapes and sizes is still challenging as the nuclei are closely packed. The watershed transform provides an efficient tool in segmenting such nuclei provided a reasonable set of markers can be found in the image. In the presence of low-contrast variation or excessive noise in the given image, the watershed transform leads to over-segmentation (a single object is overly split into multiple objects). The traditional watershed uses the local minima of the input image and will characteristically find multiple minima in one object unless they are specified (marker-controlled watershed). An alternative to using the local minima is by a supervised technique called seeded watershed, which supplies single seeds to replace the minima for the objects. Consequently, the accuracy of a seeded watershed algorithm relies on the accuracy of the predefined seeds. In this paper, we present a segmentation approach based on the geometric morphological properties of the 'landscape' using curvatures. The curvatures are computed as the eigenvalues of the Shape matrix, producing accurate seeds that also inherit the original shape of their respective cells. We compare with some popular approaches and show the advantage of the proposed method.

Contribution of diffuse inputs to the aqueous mass load of perfluoroalkyl acids in river and stream catchments in Korea.

Recent studies disagree regarding the contributions of point versus non-point sources to the aqueous mass loads of perfluoroalkyl acids (PFAAs). This study investigated the longitudinal change in PFAA mass load from upstream to downstream stations along rivers and/or streams to assess the relative contributions of point versus nonpoint inputs. With concentrations 10 to 100 times higher than running water, point sources such as wastewater treatment plants (WWTPs) effluent and airport ditch-outlet (ADO) water were separated from neighboring upstream and downstream running waters using principal component analysis. Source waters were characterized by certain predominant components [e.g., perfluorobutylsulfonate (PFBS) and perfluorooctanoic acid (PFOA) in WWTP effluent and perfluorohexylsulfonate (PFHxS) and perfluorooctylsulfonate (PFOS) in ADO water], which were minor components of running water. From a mass balance assessment of PFAA mass load, certain compounds such as PFOA and PFBS dominated the contribution of point sources to the mass load in the running water at downstream stations or in small catchment basins with high levels of industrial activity. Most of the mass load in the investigated catchments was attributable to upstream running water with a minor influence from industrial, commercial, and domestic human activities. Furthermore, the negative relationship of per capita emission factors (hereafter, EFs) with population density and a lower contribution of PFAA from WWTPs (~30% on average) compared to the running water-derived mass load at the national level indicated that diffuse inputs were more important contributors to aqueous PFAA contamination in each catchment basin as well as the entire watershed of the country (Korea). Volatile precursor compounds, which are readily dispersed to neighboring basins and transformed to PFAAs in the ambient environment, can be an important source of these diffuse inputs and will become more significant over time.