[U-04] Geospatial Applications for Societal Benefits

[U04-10] “Small and simple” keeping pace with the “big and intelligent”: Intersecting Big Data, Artificial Intelligence, and Citizen Science for smarter earth science systems

*Decibel Villarisco Faustino-Eslava¹, Jenielyn Tuando Padrones², Juan Miguel Guotana¹, Loucel Cui¹, Jefferson Rapisura¹, Francis Ian Pabillar Gonzalvo¹, Earvin Jon Guevarra¹, Gabriel Angelo Mamaril¹, Bianca Maria Laureanna Pedrezuela¹, Beth Zaida Ugat², Maria Regina Regalado¹, Rosemarie Laila Areglado¹, Donald Luna⁴, Jayson Arizapa², Carla Dimalanta³, Graciano Yumul Jr.^1,5 (1.Earth Systems Research Team Laboratory (EaRTLab), School of Environmental Science and Management, University of the Philippines Los Baños, College, Laguna, Philippines, 2.Institute of Renewable Natural Resources, College of Forestry and Natural Resources, University of the Philippines Los Baños, College, Laguna, Philippines, 3.Rushurgent Working Group, National Institute of Geological Sciences, University of the Philippines Diliman, Quezon City Philippines, 4.Institute of Biological Sciences, University of the Philippines Los Baños, College, Laguna, Philippines, 5.Cordillera Exploration Company, Inc., Bonifacio Global City, Taguig, Metro Manila, Philippines)

Keywords:Big data, Artificial Intelligence, Citizen Science, Crowdsourcing, LIGTAS

Big data requirements to generate artificially intelligent systems are common hurdles, especially for regions where data generation, collection, and archiving have only just recently become a practice. These challenges have become especially more obvious with the recent drive to make use of AI for faster and more efficient disaster risk reducing protocols, and for the forecasting, monitoring, and management requirements of smart agriculture systems. An on-going research is discussed to highlight how such difficulties are being overcome using resources that are available to most Global South countries.

Project LIGTAS or Landslide Investigations on Geohazards for Timely Advisories in the Philippines focuses providing landslide advisories based on site-specific rainfall-landslide thresholds. To generate the forecast models, extensive amounts of data, in both the spatial and temporal aspects are needed for robust statistical analyses. The database of required information in the Philippines has yet to reach critical mass. Hence, Project LIGTAS is using various approaches to build up the necessary information to move the analysis towards model generation. The use of traditional and non-traditional data collection approaches are employed to build up weather and landslide information. The growing strength of citizen-based sources has also become an integral part of the system. It has been shown to improve the robustness of systems, despite some pitfalls. This presentation is aimed at sharing experiences on big data-collecting, the challenges encountered, and how the loop between big data and AI can be improved.