CMStatistics 2020: Start Registration
View Submission - CMStatistics
Title: Seasonal variability in stream water chemistry assessed through a principal balances based approach Authors:  Caterina Gozzi - University of Florence (Italy) [presenting]
Antonella Buccianti - University of Florence (Italy)
Abstract: In modern times, climate change is significantly impacting freshwater resource availability. A warmer climate accelerates the water cycle, changing hydrological pathways, increasing evaporation, altering rainfall intensity, distribution and runoff regime. The implications on seasonal variability in stream water chemistry are significant, and the use of new compositional methods is mandatory to capture the relative behavior of solutes from a holistic perspective. A Principal Balances based approach was used to investigate the seasonal variability of 222 river waters sampled along the entire Tiber River catchment (central Italy). The same sequence of isometric log-ratio coordinates obtained for the whole dataset was applied separately to data from different flow regimes, evaluating changes due to seasonality. Principal Balances were compared utilizing a joint analysis of the kernel density distributions, and relevant deviations were highlighted using bar-plots. Results indicate a higher variability in drought periods and an influence on water chemistry of silicate weathering processes, which are strongly dependent on runoff fluctuations. The method proved to be effective and provides new insights for seasonal analysis on compositional data.