New open-source landcover time-series maps (2017-2024) focusing on India’s Open Natural Ecosystems

[M. D. Madhusudan]

Open Ecosystems, Open Data, Open Questions

Why Open Natural Ecosystems: Home to imperilled biodiversity and millennia-old pastoral traditions, Open Natural Ecosystems (ONEs) cover nearly a tenth of India's landmass—and yet, public policy disparages them as  "wastelands". These grasslands and savannas sustain a rich tapestry of pastoral and agro-pastoral communities—herders, nomads, and transhumant peoples whose languages, textile traditions, animal breeds, and ecological knowledge were forged in these open lands. From Rajasthan's sacred orans to Karnataka's ancient kavals, these commons remain vital to rural livelihoods and cultural identity. ONEs are also home to some of India's most imperilled fauna—the great Indian bustard, lesser florican, wolf, and blackbuck—and harbour a staggering diversity of endemic plants still being discovered and described. 

Paradoxically, ONEs now face pressure from short-sighted climate mitigation actions: colossal solar and wind installations to reduce carbon emissions, and mass tree-planting to sequester carbon—both sited on lands presumed to be unproductive. The ecological and  socio-cultural costs of such policies and actions deserve careful examination.  We hope our maps can help inform better decisions on the ecological  trade-offs involved in our energy and development choices.

About the Maps: Our 2025 release offers public data and tools for that scrutiny: an 8-year time-series (2017–2024)  of 11-class land-cover maps—including 6 ONE classes—for semi-arid and sub-humid India (excluding the Himalaya and NE India) at 10-meter resolution, analysed and produced on Google Earth Engine. The technical evolution of our mapping products and how their capabilities have developed over time is described in the attached PDF.

Data Access & Availability: All our maps are publicly available in analysis-ready formats on Google Earth Engine platform (starter scripts available via our web app), while our detailed methods, training data, and source code are openly available under an MIT License at https://github.com/mapping-place/one-timeseries. We share this work seeking engagement—to learn, improve, and create better public data, and to build better public understanding and policy together. 

To learn more about the latest product and our project (supported by The Habitats Trust), explore the app at https://one-india.fly.dev (best viewed in Chrome on a large screen; not optimised for mobile)

M. D. Madhusudan & Pradeep Koulgi