No abstract

Considering the current importance watercourses quality conservation, it is important to establish relationships between parameters that enable evaluation of the origins of changes in water quality, allowing actions to mitigate them. However, it is important to improve the association of different variables and to take sufficient samples. This study associates usual techniques and parameters to analyse the water quality of an urban river from Paraná State, Brazil. For this, we used biological indicators (aquatic macroinvertebrates), physical-chemical (temperature, turbidity, true colour, pH, DO and BOD5,20) indicators and microbiological (faecal and total coliforms) indicators. These indicators were related to land use and occupation classes obtained from high resolution QuickBird 2 images. For this association, the surroundings (450 meters buffer) of three distinct points of the river were considered: I. Near the spring; II. In the downtown city; and III. In a residential neighbourhood. Different values of physical, chemical and microbiological variables were detected along the river, showing evident relationships between them and with the use and occupation of the urban and peri-urban space in the characterization of surface waters. The association design was able to detect the landscape effect on water quality in a coherent way and that these connections were mainly related to suppression of the riparian forest present in the surroundings, further demonstrating the importance of this vegetation for the maintenance of watercourse quality.

Vinasse, an effluent generated during sugar and alcohol production, has great potential for soil and water pollution; however, it can be treated, used in biomass production and reused in sugarcane plantations. Thus, this work uses different types of biodigested vinasse to produce more biomass. The effect is the removal of ammonia nitrogen quickly and the end of the exponential growth phase of microalgae at different levels from the sixth day of cultivation. Among the concentrations used, the use of 50% biodigested vinasse showed the highest biomass concentration (255 mg L−1) after 10 days of growth, coinciding with the end of ammoniacal nitrogen availability and stabilization of effluent color removal. The addition of biodigested vinasse also provides an increase in chlorophyll a (5.33 mg L−1) and b (4.66 mg L−1) levels, obtained on the sixth day with 40% of vinasse, as well as protein (40.50%) with 50% effluent. Therefore, with the obtained results we noticed the variation of the biomass composition according to the vinasse concentration and increase of the pigment concentration in the presence of the effluent with higher nutrient concentration. Thus, the higher concentration of vinasse was more productive of the cultivation of Chlorella vulgaris.

& Key message We found high accuracy classification (F measure = 95%, on cross-validation) of Araucaria angustifolia (Bertol.) Kuntze, an endangered native species, and Hovenia dulcis Thunb. an aggressive, invasive alien species in WorldView-2 multispectral images. In applying machine learning algorithms, the spectral attributes mainly related to the near-infrared band were the most important for the models. & Context It is difficult to classify tree species in tropical rainforests due to the high spectral response's diversity of existing species, as well as to adjust efficient machine learning techniques and orbital image resolution. & Aims To explore the spectral and textural response of an endangered species (A. angustifolia) and an invasive species (H. dulcis) in WorldView-2 multispectral images, testing its recognition capability by machine learning techniques. & Methods We used a WordView-2 (2016) image with 0.5-m spatial resolution. Then we manually clipped the canopy area of the two species in this image using two compositions: True color composition (R=660 nm, G=545 nm, B=480 nm) and near-infrared composition (NIR-2=950 nm, G=545 nm, B=480 nm). Thus, we applied spectral and textural descriptors (pyramid histogram of oriented gradients-PHOG and Edge Filter), which selects the most representative features of the dataset. Finally, we used artificial neural networks (ANN) and random forest (RF) for tree species classification. & Results The species classification was performed with high accuracy (F measure = 95%, on cross-validation), essentially for spectral attributes using the near-infrared composition. RF surpassed the ANN classification rates and also proved to be more stable and faster for training and testing.Handling Editor: Barry A. Gardiner Contribution of the co-authors Crisigiovanni E. L. designed the methods, performed the experiments, processed the data, analyzed the results, and wrote most of the manuscript. Figueiredo Filho A. idealized the article, provided the materials (satellite image), and formulated the research framework. Pesck V. A. contributed to the geoprocessing and remote sensing analysis and cooperated in the methodology's design. De Lima V. A. cooperated in the methodology design and performed the machine learning analyses and data processing.

No abstract