The origin and research progress of Big Earth Data

Abstract

The information age has been a product of, and has been fueled by, the process of decades-long digitization, consequently transforming human societies globally. Big data is a revolutionary innovation born out of the information age that has allowed the development of many new methods in scientific research, new ways of thinking about discovery, and a new strategic high ground for knowledge economies and national interests. In the context of big data, scientific big data was proposed in 2013 to represent a paradigm shift from model-driven science to data-driven science. This transition challenges the traditional scientific research methodologies relying on observation-based science, emphasizing the search for relationships within massive datasets. This approach allows for exploration and discovery without heavy reliance on pre-existing models or prior knowledge. Scientific big data exhibits complexity, comprehensiveness, and globalization, driving a shift towards multidisciplinary approaches and international collaboration. The digital revolution is providing unprecedented access to unconventional data that quantifies societal trends at a scale not possible before. The continuous development of big data technology has brought about “Big Earth Data”, referring to big data with spatial attributes mainly generated from large-scale scientific experiment devices, detection equipment, sensors, socio-economic observations, and computer simulation processes in space and on the ground. The aim of Big Earth Data science is to significantly enhance the information value of these discrete sources of data through proper integration and interpretation. Big Earth Data provides an opportunity to fill in various gaps in data and information on global challenges, in particular sustainable development goals (SDGs), and key insights into socio-environmental interconnections. Big Earth Data, unlike conventional big data, gives geographic context and naturally emphasizes the utility of large volumes of Earth observation data and its interoperability with conventional big data systems and methods. To integrate cutting-edge science and technology in Earth, space, and information sciences, the Chinese Academy of Sciences launched its leading strategic science and technology initiative, the “Big Earth Data Science Engineering Program” (CASEarth). CASEarth paved the way for Big Earth Data to serve the development of Earth science and the United Nations 2030 Agenda for Sustainable Development. CASEarth also facilitated the establishment of the International Research Center of Big Data for Sustainable Development Goals (CBAS). An important objective of CASEarth is to scientifically understand, model, and apply the process of transforming data into information, and to provide knowledge for global sustainable development. This was a response to the important realization that SDGs have been restricted due to missing data, development disequilibrium, and association and mutual restriction between SDGs. CASEarth is working to integrate ever-increasing data sources to significantly fill various gaps in data and information on global challenges for the SDGs. These studies have been presented as a series of reports with 106 case studies covering approximately 24 targets, presenting 84 data products, 54 new methods and models, and 74 decision-support products in their respective study areas at local, national, regional, and global levels. The reports demonstrate that Big Earth Data, through its methodologies and new research perspectives, spurs revolutionary changes in Earth science and research on sustainable development. For future development of Big Earth Data, there is a need to focus on investing in infrastructure to take advantage of quickly growing digital technologies, data products for decision-making and policy for SDGs, and research into unified data standards to improve interoperability.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.