AQUASCAPE – USING ENVIRONMENTAL INTELLIGENCE TO MAP THE WORLD’S AQUACULTURE
Introduction
The bulk of aquaculture production takes place in developing nations (FAO, 2024), whereas most of the tools used to support the activity have been developed in Western countries (Ferreira et al., 2012). The consequence is an information divide that presents a major challenge for sustainable growth of the sector.
The use of mathematical models to simulate growth and environmental effects of cultivation, both for licensing and monitoring, is far more prevalent—and consigned in legislation—in countries such as Norway and Canada than in Asia and South America. The use of such tools extends to nations such as the US and various EU Member-States where aquaculture production is insignificant both from a global perspective and to their internal economies.
These issues should be a key part of the sector's agenda for sustainability over the next decades (Costa-Pierce et al, 2026).
There is widespread use of informal licensing approaches for aquaculture in many parts of the world, both on land and in coastal waterbodies—this is a problem on many levels, including assessment of accurate production data and waste footprints, biosecurity, and access to private capital. Tools based on remote sensing, machine learning, and AI can be
/>
Fig. . Fish farm clusters in an area of the South China Sea under Hong Kong jurisdiction.
leveraged on a global scale to provide solutions to some of these challenges. This paper describes the development of AquaScape (Fig. 1), a global mapping tool designed to address a number of important issues in this domain, and exemplifies how it has been applied in collaboration with the Aquaculture Stewardship Council (ASC) to improve the quality of certification of shrimp culture in India.
Methods
AquaScape (https://longline.co.uk/aquascape/) was originally conceived as a platform for communicating science to management, using a mapping metaphor to provide a simplified representation of complex simulations. This involved sophisticated scientific and technical requirements whereby the outputs of hydrological, hydrodynamic, and ecological models (Nobre et al., 2010) needed to be archived, retrieved, and synthesised on the platform—the technical challenges included accurate geospatial storage of aquaculture structures such as ponds and cages, overlaying model results using cutting-edge design, and since the platform is web-based, optimising retrieval and display speed.
Since Indonesia was the first country where AquaScape was applied, this was fundamental in bridging the information gap referred above; the various simulation tools and their outputs were displayed in the context of the world's second largest aquaculture producer (14.9 X 106 t including seaweeds, FAO, 2024), which allowed a comprehensive understanding of scaling and complexity and presented an enormous computational challenge to ensure a rich and responsive user experience.
The lack of geodata on aquaculture structures of different types was identified as a key gap for management, and it was recognised that little could be achieved in the application of more sophisticated models without a solid geographical baseline.
/>
Fig. . AquaScape registry of shrimp ponds in Andhra Pradesh, India, showing data tags for a selected production unit.
The resulting registry was stored in the AquaScape platform, which now contains about 3.5 million aquaculture structures throughout the world, corresponding to thirty million tonnes of annual production, or 32 % of world production of aquatic animals (94.4 X 106 t, FAO, 2024).
Ponds, cages, shellfish rafts, and other structures (Fig. 2) were mapped using a combination of remote sensing, human validation, and machine learning. Currently, AI plays a major role in improving data quality and helping to identify key features in the dataset that can be used to convert data into information.
AquaScape is now realising its potential as a platform for sustainability; it (i) enables informal aquaculture to be part of a registry, using a tag-based system (Fig. 2) to provide standard information such as area and volume, production status (e.g. for shrimp ponds: intensive, semi-intensive, fallow), and where applicable, lease data; and (ii) allows for models to be applied at scale to determine production, dissolved and particulate emissions, connectivity, and other key performance indicators.
An example of this kind of potential application was the connectivity assessment in Andhra Pradesh (India) executed by the ASC and Longline.
This application used data from AquaScape for shrimp pond location, complementary remote sensing data and machine learning to examine activity status, using e.g. aeration and water and vegetation colour indices, and combining these outputs with GIS multicriteria evaluation to determine suitability (Fig. 3).
/>
Fig. . Schematic representation of landscape connectivity analysis.
Synthesis of case-study outcomes
270,622 ponds were identified in the coastal districts of Andhra Pradesh using pond detection analysis, with over 97% used for shrimp production. Total activity in each district averaged 72% in 2022 (Fig. 4).
All these ponds are connected to their surrounding landscape through land, water, habitat, and climate. Farming on unsuitable land or within a 100-metre buffer zone from sensitive features (as per ASC standards) raises ecological and environmental concerns, along with highlighting significant risk to the sector due to of climate change.
Srikakulam, Sri Potti Sriramulu Nellore, and Yanam are the most promising districts in Andhra Pradesh for potential certification and further shrimp farming investment due to their favourable land conditions, limited environmental conflicts, and potential to revitalise unused ponds. SPS Nellore stands out for its ideal soil, flat terrain, and low ecological impact, while Srikakulam and Yanam offer suitable sites with manageable climate risks and minimal mangrove loss.
Fig. . AquaScape assessment of pond numbers and activity status by district in Andhra Pradesh.
East Godavari also has strong potential, though it requires careful planning to address environmental risk in some areas. In contrast, districts like Krishna, Vizianagaram, and West Godavari face significant challenges from mangrove degradation, flood risk, and regulatory complexity, making further expansion more difficult without major environmental restoration.
Remote sensing has emerged as a powerful tool for monitoring and analysing aquaculture activities and their broader environmental and socio-economic contexts. By leveraging satellite imagery and geospatial analysis, this approach provides scalable, cost-effective, and data-driven insights into aquaculture farming operations and their interactions with the surrounding landscape.
This Environment Intelligence approach can be broadly applied at scale and used to evaluate very large production areas.
Environmental Intelligence can thus be used to bridge the information divide between sustainability tools and major farming areas to improve the quality of local management and global certification.
Acknowledgements
The authors wish to acknowledge funding from the Walton Family Foundation and the Sustainable Fisheries Partnership.
Key references
Costa-Pierce, B.A., Johansen, J., Ferreira, J.G., Soto, D., Thorarensen, H.T., Msiska, O.V., Valenti, W.C., Reid, G., 2026. Rapidly Evolving Aquaculture Technologies and Expansion Need Stronger Performance Standards Based on Ecosystem Science. Ecology Letters Viewpoint, In Press.
FAO (Food and Agriculture Organization of the United Nations), 2024. The state of world fisheries and aquaculture (SOFIA). FAO, Rome, 264 pp.
Ferreira, J.G., Aguilar-Manjarrez, J., Bacher, C., Black, K., Dong, S.L., Grant, J., Hofmann, E., Kapetsky, J., Leung, P.S., Pastres, R., Strand, Ø., Zhu, C.B., 2012b. Progressing aquaculture through virtual technology and decision-support tools for novel management. In: Subasinghe, R.P., Arthur, J.R., Bartley, D.M., De Silva, S.S., Halwart, M., Hishamunda, N., Mohan, C.V., Sorgeloos, P. (Eds.), Farming the waters for people and food. Proceedings of the Global Conference on Aquaculture 2010, Phuket, Thailand. 22–25 September, 2010. FAO, Rome and NACA, Bangkok, pp. 643–704 (also available at www.fao.org/docrep/015/i2734e/i2734e00.htm).
Nobre, A. M., Ferreira, J. G., Nunes, J. P., Yan, X., Bricker, S., Corner, R., Groom, S., Gu, H., Hawkins, A., Hutson, R., Lan, D., Lencart e Silva, J. D., Pascoe, P., Telfer, T., Zhang, X., Zhu, M., 2010. Assessment of coastal management options by means of multilayered ecosystem models. Estuarine, Coastal and Shelf Science, 87, 43-62.