Abstract (EN):
Interactions between pollution sources, water contamination, and ecological integrity are complex phenomena and hard to access. To comprehend this subject of study, it is crucial to use advanced statistical tools, which can unveil cause-effect relationships between pressure from surface waters, released contaminants, and damage to the ecological status. In this study, two partial least squares-path models (PLS-PM) were created and analyzed in order to understand how the cause-effect relationships can change over two seasons (summer and winter) and how the used scale (short or long) can affect the results. During the summer of 2016 and winter of 2017 surface water parameters and the North Invertebrate Portuguese Index were measured in strategic sampling sites. For each site, it two sections were delineated: the total upstream drainage area (long scale) and 250 m (short scale). For each section, data of pressures in surface waters including point source, diffuse emissions and landscape metrics were gathered. The methodology was applied to the Sabor River Basin, located in the northeast of Portugal. In this study, it was possible to determine in which season pressures affect ecological integrity and also which scale should be addressed. The models showed the influences of manganese and of potassium concentrations in stream water on the decrease in summer water quality, while arsenic's harmful effect occurs during winter. Pastures and environmental land use conflicts were considered threats to water quality when analyzed on a long scale, whereas agricultural areas played a role when the short scale was used. The effect of landscape edge density revealed to be independent of scale or season. Effluent discharges in surface water affected the water quality during the summer season, while the effect of discharges in groundwater affected the water quality in winter. It has also been found that, to find the harmful effect of pressures, it is necessary to approach different scales and that the role of landscape metrics can also overlap contaminant sources.
Language:
English
Type (Professor's evaluation):
Scientific
No. of pages:
23