Phytoplankton silica uptake under different artificial upwelling conditions in the Canary Islands.

Climate change is affecting pelagic ecosystems with repercussions on fish production. In particular, global change is increasing oceanic temperature and stratification with decrease in nutrient input in euphotic layer leading to a decline in primary production. The mesocosm-based project Ocean Art-Up, conducted in Gran Canaria, is aimed to increase fish production and to enhance carbon sequestration through an artificial upwelling system. Diatoms dominate the phytoplankton community in upwelling systems and they need to take up silicates to grow. The abundance and nutritional value of diatoms determine the fate of phytoplankton biomass with transport to the upper level of the pelagic food web or to the deeper layer of the ocean with potential carbon sequestration. Here, data about experiments performed in 2018 and 2019 are reported. The first mesocosm experiment investigated the differences between pulsed and continuous upwelling mode, while the second experiment was conducted with a gradient in Si:N ratio along the mesocosms. The phytoplankton community takes up and incorporate silica about at the same rate in continuous mode, while in pulsed mode its peak occurred only after the deep-water addition. The diatom silica content is not affected by mode and amount of water added but by the Si:N ratio. Diatoms grown in an environment with high Si:N ratio values show higher abundance, biogenic silica production, silica uptake and silica content than the ones that experienced low Si:N values. In addition from literature, euphotic zone rich in silicate may produce high silica containing-diatoms who will produce repercussions on copepods community regarding feeding, hatching and growth, thus continuous upwelling with high Si:N ratio favours diatoms who will tend to sink and to be converted by copepods into fecal pellet rich in silica with increasing in potential carbon sequestration. Fish production may increase with continuous artificial upwelling showing low Si:N values.

Latent representations for traditional music analysis and generation

We create and study a generative model for Irish traditional music based on Variational Autoencoders and analyze the learned latent space trying to find musically significant correlations in the latent codes' distributions in order to perform musical analysis on data. We train two kinds of models: one trained on a dataset of Irish folk melodies, one trained on bars extrapolated from the melodies dataset, each one in five variations of increasing size. We conduct the following experiments: we inspect the latent space of tunes and bars in relation to key, time signature, and estimated harmonic function of bars; we search for links between tunes in a particular style (i.e. "reels'") and their positioning in latent space relative to other tunes; we compute distances between embedded bars in a tune to gain insight into the model's understanding of the similarity between bars. Finally, we show and evaluate generative examples. We find that the learned latent space does not explicitly encode musical information and is thus unusable for musical analysis of data, while generative results are generally good and not strictly dependent on the musical coherence of the model's internal representation.