JavaScript is required to use this site. Please enable JavaScript in your browser settings.

Global Database of Natural Hazards Impacts

Title: A Global Database of Natural Hazards Impacts Reported in the Scientific Literature

Duration: 01.06.2023 – present

Research Area: Earth and Environmental Sciences

In this research project, we aim to build a comprehensive global database on the impacts of natural hazards reported in scientific literature since the 1950s. To achieve this, we have initiated a systematic mapping of worldwide research on hydroclimatic extremes, including droughts, heatwaves, and floods. The corpus includes full-text open-access papers extracted from databases such as Science Direct and Pubmed. To do that, we will build a classification model to identify hazards and their reported impacts from scientific text. We will pre-train a transformer-based language model on the corpus and fine-tune it to (i) classify sentences and (ii) identify entities in the sentences describing the impacts of natural hazards, such as hazard cause, date, and location of the hazard, impacts, and number of affected people.

Aims

  • Build a corpus of 30 to 50 annotated open-access articles on the natural hazards research domain.
  • Build a peer-reviewed global database of the impacts of natural hazards, detailing date, location an causes of the hazard, and qualitative and quantitative descriptions of its social, economic, environmental and health impacts

Problem

Existing studies and databases of natural hazard impacts have several limitations, such as (1) a low level of detail on how people were affected; (2) an underestimation of the impacts; (3) a limited geographical range; and (4) a lack of information on the source of the data. However, scientific publications, reports, and handbooks compose a large data repository that can provide valuable and trustworthy information on natural hazards.

Technology

  • Natural Language Processing
  • Deep learning: text classification models based on deep learning
  • Python, R

Team

Lead

  • Dr. Taís Maria Nunes Carvalho

Team Members

  • Prof. Dr. Jakob Zscheischler
  • Mariana Madruga de Brito
  • Christian Kuhlicke

Partner

funded by:
Gefördert vom Bundesministerium für Bildung und Forschung.
Gefördert vom Freistaat Sachsen.