OR/17/045 Conclusions and next steps

Duncan, M, Mee, K, Hicks, A, Engwell, S, Robertson, R, Forbes, M, Ferdinand, I, Jordan, C, and Loughlin, S. 2017. Using the 'myVolcano' mobile phone app for citizen science in St. Vincent and the Grenadines: a pilot study. British Geological Survey Open Report, OR/17/045.

The workshop and school visits have identified that local input to modification and ownership of myVolcano are required to make it a viable and sustainable option for citizen science in St. Vincent. Citizen science is not a new activity in St. Vincent in the sense that many of the earliest observations of volcanic eruptions were made by citizens. It also has a number of institutions involved in the monitoring and managing of emergencies that would benefit from tools to share data to assist in the monitoring and management of multiple hazards and crises.

The need for an all-encompassing app, which includes multi-hazard functionality, is clear. myVolcano was not designed to be exclusively for volcanic hazards, but its name is perceived to be hindering its application beyond volcanic hazards. The list of suggested improvements in Figure 9 demonstrates the types of customisations required in St. Vincent. Whilst these customisations might also apply to other countries there are, undoubtedly, going to be additional specifications required to create an app that is fit for purpose in a particular country. There is, therefore, clearly a need for in-country agencies to have access to software that they can adapt for their own means.

In conjunction with the visit to St Vincent in 2017, there have been two parallel pieces of development work on the app. The first is the development of a cross-platform version of the app and the second is to develop more targeted multi-hazard functionality, either within myVolcano or in a new app or tool. The intention is that these tools will be made available to in-country partners, for them to make the modifications they require, manage their data locally and feed into global database of citizen science observations.

myVolcano version 2.0: cross platform and home pages

Since the workshop, myVolcano version 2.0 has been released (April 2017), which was designed based on the feedback from the scoping study (Mee and Duncan, 2015) and ongoing discussions with Caribbean partners (R. Robertson). The new version was developed as a cross-platform tool, meaning that myVolcano is now available on Android, as well as iOS. The cross-platform technology means that the app only needs to be developed once and is then ‘deployed’ onto the two different platforms, whereas previously, the development needed to be done for each of the separate platforms. The ‘app’ can also be loaded onto laptops and PCs, thereby providing a ‘desktop’ version of myVolcano that both supersedes the web version that was available during the workshop and has the same functionality as the mobile version. This new ‘cross-platform’ approach has provided a more efficient and cost-effective way of development.

Two major updates were made to myVolcano in version 2.0. The first is the improved workflow (Figure 17), which also removes the emphasis upon volcanic ash (although users can still add information about ash and see how to collect samples).

Figure 17    The new workflow for adding a record guides provided a more streamlined process for uploading observations.

The second is the addition of ‘home regions’ (Figure 18). The home region pages contain information about monitoring and emergency management in that region, and currently include 30 volcanically active countries or regions of the world. A second phase of home regions is planned, which aims to compile pages for several countries in Africa, as well as many other countries including Spain, Greece, France, and Papua New Guinea.

Figure 18    'Home region' page for St. Vincent, with links to the monitoring organisation (UWI SRC), emergency management (NEMO) and other relevant social media and citizen science initiatives.

Future versions of myVolcano will also be translated into Spanish, as with previous versions, and possibly additional languages (e.g. French, Italian, Icelandic, Amharic) determined through needs assessments with project partners in different countries. There are also plans to develop technology to enable multiple photographs to be uploaded in one record (currently only one photo per record is permitted) and to enable addition of videos (a recommendation from the pilot study).

We will also continue to look at how we can improve the range of information provided through the interactive map e.g. inclusion of hazard maps, monitoring organisation information, volcano activity alert levels etc. However, as the findings of this study suggest, more adaptations are needed to meet the requirements of users.

‘Multi-hazard’ functionality: detailed observations

Although myVolcano has been designed such that the user should be able to record an observation of any hazard (i.e. through free text descriptions), the idea behind adding hazard specific questions came as a means of capturing more detail and would also enable filtering of the data by hazard type. These more detailed questions would be an optional extra for users, who would still be guided through the initial addition of photographs or free-text descriptions.

The first proposed component of the additional questions is the inclusion of toggle boxes to categorise what type of hazard or phenomena the user thinks they have observed (Figure 19).

Figure 19    Checklist for users to identify the principle hazard they have observed.

Based on their answers, they would then be directed to additional (optional) question pages, asking for specific details regarding the hazards they have identified (e.g. Figure 3). The addition of these questions would assist with data management, since there is an early process of data organisation, making it easier for the agency receiving the data to manage and interpret the data. Categorisation will also enable observations to be symbolised by hazard type in the map, helping to improve the visual impact of observations. During the workshop, there was little opportunity to gather feedback on the proposed multi-hazard questions owing to the challenges of internet access.

Localisation and local ownership

In addition to improvements to the existing myVolcano app, the next phase of the project will look to transfer existing knowledge and technology from myVolcano to provide a more localised, decision-support tool to enable various organisations (e.g. emergency managers, monitoring scientists, emergency services and infrastructure/utility companies) to best manage and respond to multiple hazards.

The recommendations of this study are, therefore, that BGS continues to develop the technology for citizen science apps for hazard and environmental change observation, including for myVolcano and/or a new multi-hazard app, and to make this technology available to local in- country agencies. These agencies can then adapt the ‘global’ app design to meet their specific needs (co-produced with BGS if support is required), manage their local data and contribute to a global data set of citizen science observations stored by BGS. The BGS’ role would be to manage the data globally as the app is applied and adapted in other countries, thereby creating a database of citizen science observations of multiple hazards across the world, which can be used for research purposes. BGS’ legacy would be in the co-design of an extensible toolkit with partners, which could be rolled out across other countries.

Figure 20 outlines the conceptual approach and the steps to achieving these are summarised as follows:

1. BGS to co-design a generic open-source multi-hazard app that can be adapted by local agencies (a toolkit);
2. Local agencies adapt the multi-hazard app to build customised applications to meet their needs (e.g. localisation, links to social media: ‘all encompassing’ concept). NB: This could also be done by BGS if the local agencies want support in doing so, or do not have the capacity to continue development on their own;
3. Local agencies manage their own data, including the verification process and/or identify where they need data management support (e.g. from BGS);
4. Data collected at the local scales feeds into global datasets managed by BGS, as part of global initiatives, e.g. Global Volcano Model.

Figure 20    Conceptual diagram of the process of app localisation and ownership.

The plan is for BGS to support the app development through our Global Geological Risk ODA Platform, with co-design and co-development with our partners in the Caribbean and elsewhere.