Future

← Previous topic: History

An international group of 30 researchers convened at the U.S. Army Corps of Engineers Field Research Facility (FRF) 20 - 23 September 2016 for the 12th Argus Workshop. The meeting theme was “30 years of Argus at the FRF”. Afternoon discussion sessions at the workshop focused on developing a 5 – 10 year strategic plan for the Argus coastal imaging community. Workshop participants agreed that the successful long-term collection of optical coastal imaging data resulted from the automated quantitative tools developed in the Oregon State University (OSU) Coastal Imaging Lab (CIL) and discussed ways to address future challenges in collection of coastal imaging data. Themes for the future included sustainability, a move from a centralized to a decentralized network, and an overall increase in visibility of our coastal imaging community. This document summarizes the afternoon discussions at the workshop, which included the following themes:

1. Increased growth, visibility, and knowledge of the community,
2. Establishment of an International Coastline Observatory Network (ICON),
3. Development of modular, application based data products,
4. Development of a new organizational structure, and
5. Development of measures for success.

1. Increased Growth, Visibility, and Knowledge of the Community

The strategy developed at the workshop is to develop a lively and inspiring network of researchers and users that promotes and facilitates collaboration and interaction revolving around coastal imaging. The development of the Coastal Imaging Research Network (CIRN) will meet this goal through open collaboration and education. Open collaboration will be implemented by developing capabilities beyond the single user, improving documentation, creating a modern web presence, decreasing the interval between core user meetings (workshops), and training students and new users.

To improve our community, we will develop and share code using modern collaboration tools with proper version control (e.g., GitHub) and share hardware knowledge and approaches. Researchers within the network are committed to sharing innovations in software and hardware developments with the idea that more sharing will improve the overall community. The open collaboration will improve reproducibility of coastal monitoring tools and scientific results for all users. We will foster technological developments within the community by maintaining a Wiki. We will create a general Wiki to document historic innovations and discussion, share hardware developments, and maintain a “cookbook” for developing your own long- or short-term coastal imaging site. The Wiki will be especially useful to new users who wish to become involved in CIRN. Separate Wikis will be used to document specific code repositories.

The second way we will improve collaboration within our international community is through social media (e.g., LinkedIn). Social media will represent the public face for CIRN, and allow us to connect with the larger community (new users) as well as provide a useful resource for our geographically dispersed community. CIRN may also link larger earth science networks (e.g., CMSDS, OpenEarth) to improve exposure to the larger scientific community. Finally, meeting in person is an invaluable way to connect. CIRN is committed to continuing the CIL/Argus tradition of meeting every 18 months to foster collaboration. Location of the workshop is expected to rotate based on core user groups. In the future, the workshop may include calls to the broader nearshore community.

A third way to improve sustainability of our community is through education. The number of U.S. graduate students working with Argus has decreased significantly, and few students are active in other countries. We can increase the number of students and novice professional users by offering sprint session/boot camp style training aimed at fostering development of new users and developers. Initially, training will focus on producing coastal imaging products from small Unmanned Aerial Systems (sUAS). Training will include data collection techniques, data processing, and use of the cBathy algorithm. Training may include some traditional Argus “homework” given to CIL students and some discussion of basic computer vision techniques. We will organize boot camps as an extension of future workshops. Boot camps will encourage transfer of knowledge between low-level users (gurus in Argus-jargon) and high-level/non-expert users.

2. Establishment of an International Coastline Observatory Network (ICON)

Intertwined with the development of a strong community is the establishment of the International Coastline Observatory Network (ICON). These long-term coastal monitoring stations will be research-driven, high-focus, high-profile sites. ICON sites will be self-organized. Any new research partner willing to contribute a new site, or contribute to an existing site is welcome, provided the researcher/group acknowledge the collaborative model. Sites will be self-funded and must meet data storage and web hosting requirements for a robust data portal. Ideally, ICON will collectively cover a range of hydrodynamic/sedimentary environments and have world-wide representation. Requirements to join include adhering to an open source ethos. Researchers must have a shared commitment to individual site objectives as well as ICON objectives. ICON data will be open source. There will be broad research involvement including field monitoring and focused experiments at the site and a wide range of data sources other than optical remote sensing. Possible data include other remote sensing tools (e.g., lidar, radar), complementary in situ observations (e.g., tide gage, waves, bathymetric surveys), and (ideally) operational models. All sites will run standard data collections such as cBathy. Sites with an existing long time series of observations are encouraged, but this is not a requirement. Finally, the site must have the ability to disseminate images and derived products via a data portal. Sites will attempt to adhere to similar standards across ICON with a similarly themed data portal/web presence across sites. The FRF data portal (http://navigation.usace.army.mil/CHL_Viewer/FRF/) is an example of what an ICON site data portal might look like.

3. Development of modular, application based data products

Presently, CIL codes require Argus coastal monitoring hardware and rely on the Argus directory structure and database. Based on group feedback at the workshop, it is clear that a single long-term type of coastal monitoring station with integrated data acquisition and processing is not the best approach for all research questions. In the words of one attendee, the present single platform strategy is “too costly, too complicated, too specialized” for some applications. The new approach will acknowledge that different applications require different tools (Figure 1), many of which are already being independently developed. This decentralized approach is envisioned as a scalable system. Data may be across a range of optical platforms, including video from satellites or small unmanned aerial vehicles (sUAS), cameras of opportunity, or simple imaging systems (e.g., GoPro, Garmin) or more complicated short term systems and traditional long term coastal monitoring stations. We will integrate optical video with other co-located sensors including lidar, radar, and infrared cameras and will integrate with models.

The development of independent toolboxes for data processing will decouple data from the acquisition process and the Argus database/directory structure. Development of new software will focus on robust, automated, and modular toolboxes applicable to any video collected by any method. cBathy is the primary example of a robust standalone toolbox. Other examples under development are the UAV toolbox for processing video data from a quadcopter, vBar toolbox for estimating alongshore current velocity, and the toolbox for automated extraction of extreme water level.

By collaborating and sharing, developments will occur more quickly, with less reinvention. Tools can potentially be developed across the range from novice through expert users, with individuals deciding what approach best fits their application. We will create open source repositories of code. The open source ethos will increase use and development of CIRN code. Similarly, open source code will allow enterprising individuals and groups to provide services for non-expert users. As an integral part of sustainable open source code, we will adopt strict, distributed version control using Git repositories and establish tests for reliability of edited codes. We will present final data products in a universal file format (e.g., Net CDF) to simplify end user experience.

4. Development of a new organizational structure

A distributed group of CIRN leaders will coordinate and implement the changes described above. A voluntary leadership team will convene on a semi-regular basis to ensure maintenance of institutional knowledge through the enabling technologies described above and will be entrusted with keeping CIRN sustainable into the future. A group of focused teams will tackle specific challenges. A communications team will lead development of social media, collaborative coding, and education/workshops. A technology team will lead transition of CIL code to open source including addressing legal issues and developing an open license. The technology team will also foster new infrastructure development and consistency of data formats. An ICON development team will focus on long-term data collection at ICON sites, development of infrastructure at ICON sites, and work to obtain long term funding necessary to sustain ICON sites. New focused groups will be identified as needed.

5. Development of measures for success

At the workshop, we identified a number of measures for success going forward. We aim to produce a quantifiable impact on nearshore science. Impact is defined as the number of publications associated with CIRN, involvement of new principle investigators (PIs) as well as the re-involvement of PIs who have drifted from the group, and the number of new students and young professionals working within CIRN. We aim to produce a quantifiable public impact measured by press coverage of CIRN related research, number of hits on CIRN websites, and repeat visitors to CIRN websites. Other measures for success include the ability to continue long-term data collection, the development of new funding opportunities, and the engagement of applied researchers outside the core CIRN group.