Projekt

The continued expansion of slum areas to currently about 30% of the population in developing countries presents an immense challenge to urban planning capacities. It has led to the permanent retreat of the state and its public services, such as water provision, from these areas. Vulnerable populations in slums are especially at risk of experiencing highly deficient water supply and sanitation, threatening the pursuit of the UN's Sustainable Development Goal 6 ("clean water and sanitation for all"). In order to extend essential water supply services into urban slums, the society of the future could rely on AI-based Earth Observation (EO) techniques as a basis for a new planning paradigm that restores the reach of the state's planning capacity into these areas. An abundance of EO satellites employing novel sensor technologies and mission concepts are already in service or will be launched in the near future. The resulting unprecedented volumes of data have nurtured a skyrocketing development of machine learning techniques in EO (AI4EO) as a powerful and comparably inexpensive tool to address ever-more complex societal questions, such as predicting poverty risks in developing countries. Based on an exploratory case study of Pune, India, we propose to prepare an interdisciplinary collaboration that combines sophisticated AI4EO techniques with a ground-truthing approach based on socio-economic surveys into a hydro-economic multi-agent model, providing a new urban water management paradigm that will allow cities around the world to expand clean water access in urban slums.

• Prof. Dr.-Ing. Xiaoxiang Zhu

  Technische Universität München
  Fakultät für Luftfahrt, Raumfahrt und Geodäsie
  Professur für Signalverarbeitung
  in der Erdbeobachtung
  München
  

• Prof. Dr. Bernd Klauer

  Helmholtz-Zentrum für
  Umweltforschung - UFZ
  Department Ökonomie
  Arbeitsgruppe Sozialwissenschaftliche
  Wasserforschung
  Leipzig
  

• Land consumption in cities: A comparative study across the globe