We sat down together and discussed a concept. Then Dragonfly came up with some absolutely key ideas that have really made our system what it is.

David Middleton, Chief Executive, Trident Systems

Collaborative data analysis for fisheries research

Sustainable management based on robust analyses of well managed data is critically important to fishers and fishery managers in New Zealand. We support Trident Systems’ commitment to collecting better data – and getting more value from that data – to benefit the country’s fisheries.

Data collection and management

Trident coordinates the collection of catch sampling data for a range of inshore fisheries. They work with licenced fish receivers, who sample the age and length of the fish in selected landings. These data are used for stock assessments and for fisheries management.

We built a database with a web interface to receive and store the fisheries catch sampling data. The website is a key communication tool for Trident, generating email alerts, tracking sampling programme targets, and providing documentation and training resources for the fish receivers. The web interface and data entry process were designed to be as simple, error-free and reliable as possible.

Data analysis and sharing

A number of different research providers contract to Trident and analyse the data they collect. These researchers also use other fisheries data collected by the Ministry for Primary Industries.

To support this collaborative approach, we developed a code-based tool for Trident called Pisces. The tool enables multiple users to access the data but also tracks their actions via a central repository in GitHub, taking a ‘trust but verify’ approach. Users can continuously improve each other’s code, and the data owners can also inspect the code to be confident the data has been used appropriately.

Ongoing development

Before we developed Pisces, we met with Trident, talked about what they wanted and came up with a range of ideas to try out. Together we chose one to take through a proof of concept stage, which Trident then decided to progress. They have committed to its continued development in a collaborative process that will allow new research providers and databases to be incorporated in the future.

What we particularly value about Dragonfly is their combination of statistical analysis skills and IT skills. They have a good perspective on why we are doing the IT work.

The tools they build are easy to use and work well. Fisheries scientists have always had to have good IT skills, but it’s fairly unique that they are as polished as Dragonfly’s, in terms of producing IT products and website front ends.

Pisces gives us the opportunity to collaborate on analyses to a far greater extent. We feel it’s a really powerful tool for rapid data analysis. We’re doing simple analyses now that we wouldn’t have bothered with in the past because the process was too time consuming. We’re also doing the complex analyses where we rely on Pisces to ensure reproducibility.”

David Middleton
Chief Executive
Trident Systems