The Quake-Catcher Network (QCN) is a transformative approach to earthquake detection, science, and outreach. The QCN is a distributed computing project that links internal (no cost, built-in) or external (low cost, USB-based) accelerometers connected to any participating computer for earthquake research. Leveraging an innovative set of cyber-enabled seismic observations, this approach will enable the creation of a very dense, low-cost seismic network that can explore earthquake fault rupture in real-time, establish ground response to seismic wave passage, and quantify the shaking effects on critical structures. Results from a one-year exploratory grant from the NSF Cyber-Infrastructure Teaching, Education, Advancement and Mentoring program indicate QCN has the ability to be a new and transformative type of network, which is scalable and easy to deploy world-wide. Increasing the number of QCN sensors from 1,000 to more than 30,000 worldwide and developing efficient schemes to ingest, process, and distribute Terabytes of data will allow us to (1) explore fault mechanics (including directionality, slip distribution, and rupture velocity) at unprecedented resolutions, (2) study ground motions to assess seismic hazard and building response and (3) analyze data in real time for earthquake early warning and rapid response. This proposal will result in network with 6,000 new USB sensors and tens of thousands of no-cost sensors commonly built internal to laptops and other devices. Additionally, QCN will provide the cyberinfrastructure to process and analyze the large new seismic data sets in near-real time and to foster collaboration between 1000?s of researchers and interested participants around the world. The framework laid by this project will enable rapid expansion of the network internationally and will allow us to grow the network at a fraction of the cost of traditional seismic instrumentation and infrastructure, providing valuable data to augment the existing seismic networks.The success of the Quake-Catcher Network is intrinsically linked to the broader participation of the general public; members of the public, schools (K-12, undergraduate and graduate), and community organizations host QCN sensors. These ?citizen-scientists? will receive real-time earthquake information, seismic data and results, and interactive educational materials.
Through this award, we dramatically expanded the number and distribution of low-cost seismic sensors installed in homes, schools, and businesses. These sensors form the Quake Catcher Network (QCN), which takes a novel approach to increasing the density of seismic stations for improved understanding of earthquake processes and related seismic risks to the public. Several thousand sensors were distributed to members of the public in regions of the United States at high seismic hazard, and to schools, museums and other public learning centers throughout the country During the project period, we completed testing of the sensors to determine the nominal sensor response. These sensor tests are critical for both evaluating sensor quality and for use of the data in analysis of recorded ground motions. The sensors used by QCN performed as expected, producing accurate estimation of ground accelerations, albeit at low resolution than traditional, higher cost sensors. Using data collected during the project period we also pursued and accomplished several scientific studies. We were able to show that it was possible to use the low cost sensors and distributed computing techniques employed by QCN to rapidly detection earthquakes. The initial detection algorithms were developed and testing using data from an aftershock deployment following the Feb. 2010 M8.8 Maule, Chile earthquake. These algorithms were then implemented in real time during another aftershock deployment following the Sept 2010 Darfield earthquake. The results of the real time test show very rapid (8-10 sec on average) detection times and accurate magnitudes (within 1 magnitude unit) and locations (within 10 km, on average). These suggest that the low cost sensors may provide valuable real time data that could be used in earthquake early warning systems. We also completed an in-depth comparison of ground motions recorded by QCN and traditional sensors the Darfield, New Zealand earthquake sequence. Some of the QCN stations were installed close to existing network stations to examine how ground motion records compared in an ‘real-world’ deployment. We found that peak ground accelerations recorded by the QCN sensor had similar values to and similar scatter as the traditional stations. In addition, we found that when data were integrated to velocity and displacement there was little different in the time series. These results show that the QCN sensors can provide high-enough quality data for a variety of scientific studies. In addition, using the data recorded by the dense arrays installed in Chile and New Zealand we were able to estimate local site response at the station locations. Site response examines the variability in ground motions due to local geologic structure. Generally, it is difficult to get dense observations of ground motion across a particular region because seismic instrumentation is too costly and difficult to deploy. However, we installed 100 and 180 stations in Chile and New Zealand, respectively that recorded hundreds aftershocks. These data were used to examine local variation in ground motion, and in the case of New Zealand also showed evidence for liquefaction at a number of sites. Finally, we began investigating whether QCN sensors were adequate to recover building response. The response of a building dictates how a building will react to ground shaking. It is possible to estimate building response using ambient noise (wind, cultural noise) and/or small events, which can then be used to predict how a building might be have in a larger, potentially damaging event. We instrumented several buildings in Los Angeles, including the Factor Building on the UCLA campus. The Factor building response has been studied in depth using traditional instrumentation. We compared the building response recovered by QCN sensor to that previously reported and found that, while the data were noisier, we were still able to estimate the fundamental modes of the building.
