WS1: Earthquake Reconnaissance Missions in the Immediate Aftermath

We propose to conduct approximately five field investigations. Each investigation will comprise a team of international experts, two of whom will be UK academics and up to two PhD students. They will spend approximately five days in the field making observations and collecting data. We will coordinate with ​the Earthquake Engineering​ Research Institute (EERI), Geotechnical Extreme Events Reconnaissance (GEER) Association, and New Zealand Society for Earthquake Engineering (NZSEE) to determine whether they are mounting a mission, what features they are particularly interested in, and what contacts they have in the region.  A team leader and base-co-ordinator will be appointed, with the team leader, Principal Investigators (PIs) and  Earthquake Engineering Field Investigation Team (EEFIT) management team then tasked with selecting a team with appropriate experience and expertise. The main differences between missions to be undertaken in this proposal compared to the previous are that: (1) team members will have been educated and have strong familiarity with a pre-determined set of data collection tools (WS5), (2) the data collected will be supplemented with other primary and secondary information sources and will be more accessible to other researchers and stakeholders through the (Spatial Data Infrastructure) SDI created in WS4, (3) the reconnaissance findings and associated research will be more widely disseminated through traditional means, educational videos shot in the field, and through our network of partners.

Deliverable: Five earthquake missions and associated collected data.

Responsible people: Dr Sean Wilkinson, Prof Dina D'Ayala, Dr Emily So, Dr Claire Ellul and Dr Diana Contreras.

WS2: Return Missions and Recovery Data Collection

Developing nations still have low uptake rates of modern earthquake protection and mitigation programmes and, even post-event, the tendency is for these communities to continue to practice traditional construction practices. While there is still much to learn in terms of structural deficiencies in the built environment, we propose that the greatest gains in transforming the resilience of vulnerable communities can be made by understanding the impediments to "building back better". For this reason, we will conduct approximately three return-missions to areas affected by earthquakes. These will be selected from sites of previous EEFIT reconnaissance missions and are aimed at collecting information on the earthquake recovery process. This will include the study of recovery policies, the process of providing temporary to permanent housing, casualty and people management. Longitudinal studies of earthquake reconstruction and recovery are almost non-existent, and yet they are fundamental elements of all theorised resilience frameworks (e.g. UNISDR's Sendai Framework). Hence, to advance the study of resilience in practice, it is essential that we collect case-study data that can provide insights into how communities recover from earthquakes and how investments implemented after the recovery phase influence their resilience in the short and long-term. Return missions conducted under the auspices of this grant will build on previous return missions, with at least one of the return missions deployed to the site of either the 2009 L'Aquila Earthquake or 2010 Japan earthquake and tsunami, to provide a long-term recovery perspective. Other potential sites are areas affected by the 2015 Nepal and 2016 Ecuador earthquakes.​​ We will deploy a multi-disciplinary team, including EEFIT members with expertise in urban planning and disaster management in addition to the traditional engineering disciplines. Where possible, we will conduct in-field training of engineers without reconnaissance experience. It is therefore likely that team size will be larger than those deployed as part of WS1.

Data types required to explain how events unfold during earthquake recovery encompass sources that are different from those collected immediately after a mission. For example measures of the performance of infrastructure and other services, insurance pay-outs, finance, and labour availability, as well as social barriers. Hence, significant effort will be required to explore these data types (using the experience from the previous return missions) and ensure they are integrated within the data protocols of WS3. These return missions will also act as testbeds for the developed SDI in WS4 and international reconnaissance teams (e.g. NZSEE and EERI) and attempt to conduct joint return missions, where there is mutual interest.

Deliverable: Three return missions and collected data.

Responsible people: Prof. Tiziana Rossetto, Dr Emily So,  Prof. Dina D'Ayala, and Dr Claire Ellul.

WS3: Data Collection Protocols for Effective International Dissemination

EEFIT missions have been instrumental in collecting post-earthquake impact data for over 25 years, and its archives contain unique and valuable information for seismic risk modelling purposes. This data has been used to develop building inventories, conduct vulnerability assessments, and to calibrate and validate catastrophe models. However, it is increasingly clear that understanding the entire hazard chain is necessary for assessing consequences for communities. Data protocols for building damage due to ground shaking were developed in the previous grant. Within this proposal, we will extend this to consider earthquake secondary hazards (such as landslides) as well as social consequences.

Post-earthquake data to be included in the review and development of new data protocols and for inclusion in an SDI (See WS4) will include: a) ground shaking damage to buildings; b) human casualty and consequence studies and statistics; c) ground shaking effects on critical facilities, infrastructure, and lifelines; d) consequences due to secondary hazards (landslides, liquefaction, tsunami, and fire); e) socio-economic consequence and recovery data.

In this WS, the focus will be on the standardisation of new forms of global data, setting a global standard in collaboration and consultation with international partners like the Global Earthquake Model (GEM) Foundation in recording disaster losses and (Integrated Research in Disaster Risk) IRDR's DATA Project, and our other post-earthquake reconnaissance mission partners (e.g. EERI, GEER, GNS). We shall address each category of data separately and provide examples of how consequence data can be collected and communicated efficiently. This is to ensure uniformity in the data to be included in the EEFIT SDI (See WS4), and with our partners worldwide so that information can be shared and understood. 

Deliverable: an international set of data collection protocols.

Responsible people: Dr Emily So, Prof. Dina D'Ayala, Dr Claire Ellul and Miss Enrica Verrucci.

WS4: Design and Development of a SDI: for data management and sharing 

To ensure the standardised handling of the collected data and to provide a platform for better dissemination and analysis and, in turn, decision-making, we will develop an SDI. An SDI brings together a framework of spatial (mappable, geo-information) and non-spatial data, metadata, users, and tools in order to share and use this data. We will design our SDI to capture the full spectrum of earthquake impacts and data, along with additional data from third-party sources. It will also incorporate previous efforts undertaken towards data sharing by EEFIT and other organizations that have collected valuable post-earthquake datasets that can be federated. These include data in the Virtual Disaster Viewer tool, which also contains data from previous missions (the CAR and the GEMECD database)​. Data will be uploaded to the SDI from the field (or shortly after mission return) and in parallel to the production of the mission reports. This will ensure that valuable raw data and information from surveys and interviews (that are commonly either aggregated or omitted from standard EEFIT reports) are not lost. The SDI will also ensure that EEFIT meets the research data management requirements of the Engineering and Physical Sciences Research Council (EPSRC).

The core strength of an SDI is the ability to integrate very disparate data based on location. While this is already achievable in 2D, the EEFIT project will also provide an opportunity to explore the inclusion of 3D and even 4D (time) datasets, as well as to explore the potential of 3D location (e.g. 3D models of buildings) as an integrator of earthquake, structural and demographic data. This can act as an index to facilitate data organisation and search, e.g. by attaching a photograph or a laser-scan of a structural issue to the specific location on the 3D building rather than just associating it with the base footprint, facilitating queries such as 'show me all the damage that occurred above ground floor level'. To this end, a User-Centred Design (UCD) approach will be taken for the construction of the SDI - end-users' opinions will be elicited beforehand to gauge needs and expectations and to design characteristics of the expected working environment. The SDI will be developed as a 3-tier web-based architecture (database/repository, web server, end-users) for access not only by desktop PCs but laptops, tablets and mobile phones. To ensure the long-term sustainability of the data repository and to minimise license costs, well-established off-the-shelf software packages Free and Open Source (FOSS) software will be used. These tools will also help to ensure data interoperability. Relevant metadata will also be reviewed to identify or develop appropriate standards for data documentation which maximises description while minimising cognitive load to understand the description, and the potential for its automatic collection (linking closely to protocols developed in WS3 and tools from WS5) examined.

The repository and this WS will act as an integrator and end-point for the other WS in the project, as well as a store for project data, underpinning EPSRC data management requirements. The SDI will be launched in a staged approach, with a preliminary version released 12 months into the project. Annual releases of new versions will contain new developments and modifications in response to feedback.

Deliverable: an SDI for storage of and access to field mission data.

Responsible people: Dr Claire Ellul, Dr Philip James, Prof. Tiziana Rossetto and Miss Enrica Verrucci.

 WS5: Tool Development and Workshops

Data capture technologies have been advancing at an exponential rate. Smartphones and apps, 3D imaging and terrestrial laser scanning, as well as (unmanned aerial vehicle​) UAV technology, has enabled the rapid capture of data and miniaturisation has enabled sophisticated engineering equipment to be deployed in the field. In addition, social media has the potential to dramatically increase both data collection and data interpretation communities. This project will produce a new generation of data collection tools as well as provide pre-mission experience in the use of these tools. Rapid Visual Surveys (RVS), aimed at determining damage distribution or acquisition of basic typology data will be conducted by the omnidirectional camera and UAV scanning integrated on tablet supported data collection forms developed according to the data collection protocols defined in WS3. A specific App will be developed to extract relevant data from the imagery and populate the digital GIS referenced databases of the SDI developed in WS4. These tools will have three purposes: while people are on site, decisions can be made and communicated on specific areas of investigation and data collection, as new information can quickly be acquired and shared with all mission members; during end of day debriefings they can be used as support for decision making to decide on the next day's programme. They will also be used in first and return missions to provide repeated takes of the same visual data to determine the pace and extent of recovery, hence linking WS1 and WS2. These digital databases will be integrated with more technical survey format such as the already available EEFIT version of the GEM ICDT Android application to include the FAMIVE and Log-IDEAH damage to masonry buildings and a consequence data collection form.​

Additionally, we will investigate Social Media (SM) as an alternative way of rapidly collecting large amounts of damage information for the SDI through tweeted and uploaded images and videos of damaged buildings. Additionally, we will develop a simple twitter assessment protocol trialled for damage assessments by novice users as well as interpretation of 3D omnidirectional images collected in the missions. Methodologies to assess the quality of novice data will be developed e.g. comparing social media assessments with those made by EEFIT experts using virtual reality and if necessary calibration of novice data or weightings attributed to observations based on expertise. Social Media has already been used extensively in Emergency Response and Disaster Relief and Management and been demonstrated as an effective means to gather information about the location and severity of events and by USGS who will work with us on this system. We will assess data from Twitter and Instagram for its ability to monitor impacts and the effectiveness of relief operations. For example, keyword monitoring will be used to automatically request contextual information or images from twitter users. Using this information we can start to obtain causative relationships between failures in the built environment and community impacts and measure recovery rates. Other equipment (e.g. microtremor and laser scanners) will be assessed for their ability to provide data for post-earthquake engineering studies. 

Deliverable: a range of tools for data collection disaster monitoring and an assessment of their potential in Disaster Risk Reduction (DRR).

Responsible people: Dr Philip James, Prof. Dina D'Ayala, Dr Sean Wilkinson, Dr Claire Ellul, Dr Diana Contreras and Miss Valentina Putrino.