In the face of aggrandizing climate change and increasingly extreme precipitation patterns, landslide hazards are expected to increase in the near future. Wide-area rainfall monitoring will not be sufficient for landslide early warning; we need a better understanding of the local dynamics of the landslide-prone area for a better judgment on a reliable warning. The current landslide monitoring instruments (e.g., inclinometer, total station, airborne laser altimetry techniques) can only provide information on the static movememt characteristics of the slope. However, detailed knowledge of the kinematics and dynamics ( e.g, microseismic activity) of slopes is essential for the understanding and prediction of catastrophic slope failures. However, large-scale, continuous and in-time...[ Read more ]

In the face of aggrandizing climate change and increasingly extreme precipitation patterns, landslide hazards are expected to increase in the near future. Wide-area rainfall monitoring will not be sufficient for landslide early warning; we need a better understanding of the local dynamics of the landslide-prone area for a better judgment on a reliable warning. The current landslide monitoring instruments (e.g., inclinometer, total station, airborne laser altimetry techniques) can only provide information on the static movememt characteristics of the slope. However, detailed knowledge of the kinematics and dynamics ( e.g, microseismic activity) of slopes is essential for the understanding and prediction of catastrophic slope failures. However, large-scale, continuous and in-time landslide monitoring is considered impractical due to the constraints of equipment and maintenance cost. Here we propose GeoSAIL (Geotechnical Scalable Architecture for In-time Landslide Early Warning and Monitoring) to fulfill the emerging needs of a high sampling rate and continuous landslide monitoring solution which is scalable and flexible in cost and performance of a large-scale deployment. GeoSAIL is a data-driven early warning and monitoring architecture re-thought from the ground up, as it: (1) utilizes consumer grade commodity off-the-shelf (COTS) hardware; (2) increases reliability by adding multiple redundancies since COTS hardware are truly economical ; (3) achieves fast development to deployment cycle through a hardware-independent sensor node architecture; ( 4) empowers flexibility in hardware configuration to facilitate site-specific alterations, and (5) leverages the open source big data ecosystem for scalable data storage and processing.

Open Smart Soil Particle (OpenSSP), an open source implementation of GeoSAIL architecture is then presented. Through OpenSSP, the effectiveness and performance of GeoSAIL are evaluated; as of now, there are as many as 11 OpenSSPs deployed in the monitoring of the Lushan and Alishan landslip area in Taiwan. Through the data collected from both consumer and seismic grade sensors installed on site, we demonstrate that the low-cost consumer grade Micro-Electro-Mechanical Systems (MEMS) sensors had comparable performance to seismic grade sensors in strong motion sensing (sensitivity (> 0.24 gal) and low frequency range (> 0.5 Hz). Such performance grade provides sufficient competency in the detection of anomalies in vibrations and movements triggered by a landslide or an earthquake. The current OpenSSP implementation is scalable in terms of cost (only USD$100 per node) and performance (no upper limit in the number of OpenSSPs installed). Furthermore, the flexibility of GeoSAIL/OpenSSP empowers the practitioner to substitute any sensor and hardware component easily, making GeoSAIL/OpenSSP extremely adaptable in the monitoring of diverse types of flow landslides (from debris flow to slope creep). In the age of climate change, better predictability of landslides can only be achieved with large-scale, real-time and long-term collection of field data. By embracing the paradigm of Big Data at the heart of the design, we aspire for a better preparedness and resilience in landslide risk and management against climate uncertainties.