Search

Clandestine connections represent a significant portion of the apparent water losses. The objective of this research work is to propose a systematic approach to inspect sites with possible illegal water consumption more precisely. A case study was conducted in a District Metered Area (DMA) located in the city of Joinville, Southern Brazil, using technical information and consumption history from 1400 consumer units located in this area. The proposed methodology uses the bagging technique called Random Forest, through a machine learning algorithm. With success rates of 88.10% when classifying frauds in the model training phase and 88.20% in the test phase, the obtained model shows great ability to properly classify frauds in the water supply system in the studied DMA. The use of the random forest classification model, combined with the SMOTE technique for data balancing, proves to be a viable technical alternative for the detection of unauthorized water consumption. [ABSTRACT FROM AUTHOR]

Changing hydrological processes and soil characteristics are of major environmental concern in surface mining. Runoff may be the best factor for determining the efficiency of rehabilitation after this type of mining, and needs more in‐depth studies, but the environmental factors require more attention in order to control that runoff. The objective of this study was to compare the runoff during post‐mining rehabilitation after the extraction of bauxite with that before mining, and to identify the factors that most affect it. Runoff from before the mining operations and during the rehabilitation after those operations were collected in seven plots in the area of a bauxite mine in southeastern Minas Gerais State, Brazil. The physical‐environmental and anthropic factors interfering with the runoff were characterized for both periods, and their results were submitted to principal component analysis to define the parameters, which most demand attention in the mining rehabilitation stage. This study provides the first records of runoff in the field in areas rehabilitated after surface mining in Brazil, where the runoff decreased (1.75‐times or 42.79%) from 0.76 to 0.30% of the precipitation during rehabilitation, compared to the reference area. The total porosity, density, and resistance of the soil, associated with the maintenance of slope, surface roughness, and vegetation cover, are the factors demanding more attention during the rehabilitation stage of surface mining in order to minimize runoff. [ABSTRACT FROM AUTHOR]

No‐till (NT) is a conservation system that improves the hydrological regime of agricultural slopes by providing greater surface protection and benefits to the physical and hydrological properties of soils. However, the isolated use of NT is not enough to control runoff and its associated degradation processes. Therefore, this study aimed to evaluate the runoff of agricultural slopes under NT under different runoff control conditions by monitoring 63 rainfall events in two 2.4‐ha zero‐order catchments and 27 rainfall events in four 0.6‐ha macroplots. The catchments are paired and similar in terms of the type of soil and relief, but different regarding the presence of terraces. The macroplots have different soil and crop management systems. By using monitoring techniques, the hyetographs and hydrographs revealed the influence of the different types of management on the catchments and macroplots and allowed rainfall characteristics, runoff volume, runoff coefficients, water infiltration, peak runoff, response times, and curve number to be analysed. The terraces positively affected the NT and controlled runoff and related variables, in addition to infiltration significantly increasing and runoff reducing in the terraced catchment. All the hydrological information assessed pointed to the positive effects provided by the presence of the terraces. The results in the macroplots showed that high amounts of phytomass and/or chiselling do not control runoff and its correlated variables in medium and high magnitude events. The study concludes by underlining the need for additional measures to control runoff (terraces), even in areas under NT and with high phytomass production. Additionally, the study emphasizes the importance of monitoring at the catchment scale to better understand the hydrological behaviour of agricultural areas and provide the necessary parameters to effectively control runoff. [ABSTRACT FROM AUTHOR]

In semiarid areas producing corn in the State of Paraiba (Brazil), despite the social and economic importance of this culture associated with some types of stress mainly in semiarid regions, as well as the soils of these regions, which in many cases have a low organic matter content, generally < 1.5%, limiting the viability of the crop. The correct management of organic fertilization via fertigation is a possibility to minimize water losses by evaporation. The objective was to evaluate the vegetative growth and productive parameters of corn (Poaceae) fertigated with bovine biofertilizer under irrigation levels. The experiment was carried out under field conditions, in the locality of the agroecology sector (UEPB, Catole do Rocha, Paraiba, Brazil). The treatments were distributed in blocks at random with four replications, in a 3 × 5 split plot scheme, represented by three irrigation depths (60, 90, and 120% of ETc—crop's evapotranspiration) that when applied daily were considered the main plots and the subplots represented by the doses of biofertilizer (0, 20, 40, 60, and 80 mL plant−1) applied biweekly via fertigation, totaling to 60 experimental plants. Plants on water stress conditions inhibit their growth and production, but when associated with biofertilizer, they result in plants with greater productive potential in semiarid conditions. The blade of 120% of the ETc provided greater growth and production of corn compared with the layers of 60 and 90% of the ETc. The 50 ml plant−1 biofertilizer dose provided corn growth and production compatible with conventional cultivation. [ABSTRACT FROM AUTHOR]