Cross-lingual access to textual evidence and impacts of weather and climate extreme events

Weather and climate extreme events cause a large number of human and material losses, as well as impacts on different sectors of society and on the environment. Therefore, mitigation and adaptation to them are a priority for national and international agencies, particularly in the current context of climate change, which is expected to increase the occurrence and magnitude of extreme events in many regions worldwide. Currently, a wide range of tools exists to measure the magnitude of extreme events from meteorological, climatic or hydrological perspectives. However, studying their impacts still presents major challenges, as most international databases cover only short time periods, do not encompass the full range of extreme event types or impacts, and tend to focus primarily on high-impact events. This makes them insufficient for detailed studies. However, there are a large number of documentary sources (operational bulletins, technical reports, post-event analyses, newspaper articles, or scientific papers, among others) that provide unstructured information, which is reliable and in some cases authoritative,  on the characteristics of the event and its impacts. In many cases, these documentary sources are presented in the local language of the country in which the event occurred. These two barriers (unstructured information and the language) clearly limit access to invaluable information on the impacts of extreme events. The CLASiK project (Cross Lingual Access to Scientific Knowledge) addresses these barriers; its main objective is to facilitate smooth and interoperable access to multilingual documentary sources, data hubs and repositories with information about impacts of extreme events on the Web for stakeholders and researchers who engage with them in their own native languages. To that end, we will build methodologies and tools for extracting, indexing and accessing scientific documents in several languages, making them interoperable and accessible to query in natural languages. The specific objectives of the Project related to this case study are: i) To provide researchers and stakeholders with a reliable framework that helps solving discovery of information on extreme events impacts across different languages. This framework will be demonstrated by developing a unified knowledge graph and Web portal to access different mono/multilingual data sources; ii) To develop a family of language services and tools that make monolingual data silos interoperable through a multilingual knowledge graph, semantically annotate scientific data and documents to extend such knowledge graph, and allow the users to query such multilingual repository and translate the answer back to their own language; iii) To disseminate and communicate the project results according to open science principles, to raise awareness, exchange of ideas with researchers from outside the consortium, and build a community of stakeholders.