T5 – Exploring spatiotemporal dynamics of sediment transport in the Zambezi River Basin

Short description

Sediment connectivity is a fundamental property of river networks, emerging from temporal and spatial interactions among sediments of different composition and grain size. These sediments are delivered into the river from sources characterized by varying supply rates, timing, and spatial distribution. The transport processes of sediment in rivers are directly linked to fluvial processes and ecosystem services, and they are highly affected by river anthropic alterations, such as dam and land use changes.

In this context, it is crucial to characterize basin-scale sediment connectivity to enhance our ability to quantify potential future alterations resulting from anthropogenic disturbances, whether direct (e.g., reservoir construction and management) or indirect (e.g., land use change, climate change). Understanding and modeling network-scale sediment connectivity and its response to anthropic alterations are key for gaining insights into river processes and informing effective river basin management.

This thesis will primarily rely on D-CASCADE, a dynamic, network-based modelling framework designed for analyzing sediment transport and connectivity in large rivers (1), (2). D-CASCADE provides estimates of sediment fluxes and reach sediment budgets throughout the entire river network, enabling the tracking of sediment position and movement over time, as well as across space.

The aim of the thesis is to configure D-CASCADE for the Zambezi River basin, which stands out as one of the most heavily dammed rivers in Africa, featuring more than 30 large dams that control its flow and sediment transport dynamics (3). The goal is to employ the model to investigate the spatiotemporal evolution of sediment supply and delivery across the Zambezi River basin. This exploration involves evaluating the effects of multiple heterogeneous drivers of change in river sediment dynamics, including dam construction and operation, as well as changes in land use.

To achieve this goal, we expect the student to carry out the following activities:

  • Literature review: understanding the fundamental operational principles of CASCADE and D-CASCADE models, as well as their input requirements; reviewing and understanding the different fields of applicability of these models; reviewing and analysing the key features and challenges of the Zambezi River basin to contextualise the framework within the case study.
  • Data elaboration: Retrieving and manipulating morphological and hydrological (either observed or derived from a hydrological model) input data for the spatial and temporal domains relevant to the case study.
  • D-CASCADE simulation and results interpretation: investigating the spatiotemporal evolution of sediment supply and delivery across the Zambezi River basin, evaluating the effects of multiple heterogeneous drivers of change in river sediment dynamics, with the aim of inferring relevant conclusions.

References

  1. Marco Tangi, Rafael Schmitt, Simone Bizzi, and Andrea Castelletti. ‘The CASCADE toolbox for analyzing river sediment connectivity and management’. Environmental Modelling & Software, Volume 119, 400-406, September 2019. DOI:  https://doi.org/10.1016/j.envsoft.2019.07.008
  2. Marco Tangi, Simone Bizzi, Kirstie Fryirs, and Andrea Castelletti. ‘A Dynamic, Network Scale Sediment (Dis)Connectivty Model to Reconstruct Historical Sediment Transfer and River Reach Sediment Budgets’. Water Resources Research, volume 58, no. 2, 25 January 2022. DOI:  https://doi.org/10.1029/2021WR030784
  3. Nicolò Stevanato, Matteo Rocco, Matteo Giuliani, Andrea Castelletti, Emanuela COlombo. ‘Advancing the representation of reservoir hydropower in energy systems modelling: The case of Zambesi River Basin ’. PLOS ONE 16, no.12, 2 December 2021. DOI: https://doi.org/10.1371/journal.pone.0259876

Relevant courses and knowledge

Natural Resources Management

Number of students

1

Requisites

The student should be comfortable with data analysis and machine learning tools. Proficient coding skills in at least one of Matlab and Python are mandatory.