Large-scale mapping of marine and coastal ecosystems: opportunities, gaps, pitfalls and the future

Perhaps the earliest systematic map of any marine ecosystem was Charles Darwin’s 1842 coral reef map. Over the next 140 years, updates and other ecosystems followed, albeit at a slow pace until we entered the digital era. The rate of mapping picked up with opportunities for digital data compilation and storage, while the onset of high resolution, low-cost remote sensing accelerated these advances. AI is heralding another step change, adding new dimensions in image interpretation and data enhancement. As the base maps have developed, a second wave of science has sought to map other dimensions: threat and condition, change over time, and multiple “values” around biodiversity and ecosystem services. Such maps, at the largest scales, are a core component of conservation planning, framing policies and actions and driving practical management including protected areas designation and marine spatial planning. Despite the continuing challenges of “seeing” through the water column or distinguishing diverse ecosystems in remote locations, we are now at a point where numerous maps exist, where duplication is beginning to occur, and where there are both synergies and contradictions in what we think we know.

We will present a synthesis of the state of coastal and marine ecosystem mapping across 20 habitat classes at large-scales – including kelp beds, hydrothermal communities, sandy shores and coral reefs. We will also describe progress on concomitant mapping of biodiversity and ecosystem service values, highlighting some of the many platforms and places where such data can be viewed, interrogated and often downloaded. Considerable gaps still exist in our knowledge, notably around deeper waters, and around dynamic, ephemeral or seasonal ecosystems. Opportunities also abound to do more with what we do know, for example around the connectivity across ecosystems and the interacting and combined values that are generated between them. Multiple avoidable challenges exist, including those of duplications and missed opportunities. At the same time this is an opportune moment to optimise investments and to maximise the utilisation and influence of both data and the learning that lies behind it. Advancing collaboration into the future may include enhanced data-sharing; establishing best practise approaches; improving inter-operability between datasets and platforms; and agreeing on a common language and definitions.