Even as the world grows increasingly reliant on data and artificial intelligence, concern over the environmental impact of data-related activities is increasing. Solutions remain elusive. The rise of generative AI, which rests on a foundation of massive data sets and computational power, risks exacerbating the problem.

In the below, we propose that responsible data stewardship offers a potential pathway to reducing the environmental footprint of data activities. By promoting practices such as data reuse, minimizing digital waste, and optimizing storage efficiency, data stewardship can help mitigate environmental harm. Additionally, data stewardship supports broader environmental objectives by facilitating better decision-making through transparent, accessible, and shared data. In the below, we suggest that advancing data stewardship as a cornerstone of environmental responsibility could provide a compelling approach to addressing the dual challenges of advancing digital technologies while safeguarding the environment.

The Environmental Costs of Data

The environmental costs of data-related activities are now apparent. In 2022, data centers already consumed 460 TWh of electricity, 2% of global usage. Cooling systems alone account for up to 40% of a data center’s energy consumption. This demand is projected to double by 2026, due to the rise of AI and other resource-heavy technologies.

The environmental burden goes beyond the well-known difficulties of data centers, however, extending downstream to the production of devices such as satellites, sensors, and other hardware and devices, all of which depend on rare earth materials such as lithium, copper, and cobalt. Mining for these materials leads to habitat destruction, biodiversity loss, and water scarcity, as seen in Chile, where mining operations exacerbate drought conditions and create conflict with local Indigenous and agricultural communities. These operations also generate large amounts of toxic waste, including tailings, which pose risks such as water contamination for local communities.

In addition, data can add to environmental burdens in unexpected ways. Take the use of satellites. While the energy-intensive processes of rocket launches and power requirements for satellite operations contribute to substantial carbon emissions, the increasing data collection capabilities of satellites exacerbate the situation. Indeed, the various energy requirements of satellite systems rival the lifecycle emissions of some small nations. Finally, satellites generate space debris at the end of their lives, adding to sustainability concerns.

As indicated earlier, data processing and storage infrastructure, such as data centers, magnify the environmental strain. In 2022, data centers already consumed 460 TWh of electricity — 2% of global usage — and this demand is projected to double by 2026 due to the rise of AI and other resource-heavy technologies (see also figure below from UNCTAD’s Digital Economy Report 2024). Cooling systems alone account for up to 40% of a data center’s energy consumption.

Data Stewardship as Environmental Stewardship

Considered together, these interconnected challenges emphasize the urgent need for sustainable practices in data-related activities to mitigate environmental and social impacts.

Data Stewardship, focused on the re-use of data in a systematic, sustainable, and responsible way, can help achieve many of these goals. To understand how this is so, consider the multi-pronged nature of the concept–i.e., the various tasks and responsibilities undertaken by data stewards. These include:

Optimizing Data Reuse to Reduce Resource Footprints: Data Stewardship promotes data minimization and the systematic reuse of existing datasets, reducing the need for resource-intensive new data collection. This cuts down on the energy, water resources, materials, and infrastructure often required to gather and process new data. One example: Re-using satellite imagery for environmental monitoring instead of launching new satellites can conserve resources and reduce emissions.

Promoting a Circular Economy for Data (Circular Data): By establishing frameworks for data sharing and interoperability, Data Stewardship encourages a “circular economy” for data. Data collected for one purpose (e.g., weather forecasting) can be reused for others (e.g., climate change modeling), ensuring its value is maximized without additional environmental costs.

Improving Efficiency in Data Management: Responsible stewardship involves fit-for-purpose analytics, optimizing storage, minimizing duplication, and using energy-efficient infrastructure (green computing). This reduces the environmental costs associated with maintaining large-scale data storage and processing systems. It also promotes sustainable digital practices by prioritizing compact data formats, reducing energy consumption in computational tasks, and leveraging scalable, green technologies.

Aligning with Environmental Goals: Many principles of Data Stewardship — such as data re-use, transparency, and accessibility — enable better environmental decision-making and help mitigate the effects of climate change. For example, shared data can inform sustainability policies, resource conservation strategies, and carbon footprint reduction efforts.

Reducing Digital Waste: Poorly managed data can lead to “digital waste” (e.g., unused or redundant datasets consuming storage resources). Data Stewardship ensures that more valuable, well-documented, and accessible data is retained and reused, reducing unnecessary energy consumption in storage systems.

Data Stewardship as Environmental Stewardship (Produced by Authors using Napkin.ai)

Call to Action and Further Research

Data Stewardship is more than a technical or data governance practice — when done responsibly, it’s also a commitment to environmental responsibility. As the demand for data grows in our digital and AI age, embedding sustainability into the data stewardship agenda is essential.

Positioning Data Stewardship as Environmental Stewardship allows us to align data-based policy and innovation with sustainability goals, minimizing the environmental impact of data-based activities while harnessing their environmental and societal benefits

Operationalization

To operationalize the vision of data stewardship as environmental stewardship, several steps will be needed including:

1. Develop Practical Guidelines for Environmental Data Stewardship: There is an urgent need to create clear, actionable guidelines that enable data stewards to integrate sustainability into their practices. These guidelines should cover areas such as optimizing data reuse, reducing digital waste, and adopting green computing standards. They should also account for sectoral differences, recognizing that environmental challenges and data-related activities vary across industries.
2. Position Data Stewards as Strategic Actors: Organizations must elevate data stewardship to a strategic level, recognizing data stewards as key players not only in increasing the value of data but also in driving sustainability. This includes embedding data stewardship roles within organizational leadership, equipping stewards with the tools and authority needed to influence decision-making, and fostering a culture that prioritizes sustainability.
3. Adopt and Incentivize Circular Data Economies: To maximize the value of existing data while minimizing environmental impacts, organizations should adopt frameworks that promote data sharing, interoperability, and reuse. Policymakers and industry leaders must work together to create incentives — such as tax benefits, grants, or certifications — that encourage organizations to participate in circular data economies.
4. Integrate Sustainability Metrics into Data Governance Frameworks: Organizations should measure and report on the environmental impact of their data-related activities. This includes tracking energy consumption, carbon emissions, and resource usage associated with data storage, processing, and collection. Such metrics can guide more sustainable practices and align data stewardship with broader environmental goals.
5. Foster Cross-Sector Collaboration: Addressing the environmental challenges of data requires collaboration between the public, private, and civil society sectors. Partnerships should focus on sharing best practices, aligning sustainability objectives, and pooling resources to tackle common challenges, such as the energy consumption of data centers or the lifecycle impacts of AI technologies.

Research Agenda

While these recommendations provide a roadmap for action, significant gaps in knowledge and practice remain. To support the transition to sustainable data stewardship, we propose the following research priorities:

1. Developing Environmental Metrics for Data Stewardship: Research is needed to identify and standardize metrics that capture the environmental impacts of data-related activities. This includes understanding the carbon footprint of data processes, water resources needed for data centers, the lifecycle impacts of data infrastructure, and the benefits of data reuse.
2. Exploring the Role of AI and Emerging Technologies: Future research should investigate how AI and other emerging technologies can be harnessed to optimize data processes and reduce environmental impacts. This includes exploring lightweight AI models, decentralized data systems, and renewable-powered computing solutions.
3. Assessing Barriers to Circular Data Economies: Studies should examine the practical, cultural, and technical barriers to adopting circular data economies, including issues of data interoperability, trust, and governance. Identifying these barriers will help design targeted interventions to promote data sharing and reuse.
4. Understanding the Social Implications of Environmental Data Stewardship: More research is needed on how environmental data stewardship intersects with issues of equity and environmental justice, particularly in under-resourced regions or communities disproportionately affected by environmental degradation.
5. Guiding Organizational Change: Research should focus on how organizations can effectively integrate environmental considerations into their data stewardship practices, including the role of leadership, organizational culture, and policy frameworks in driving change.
6. Documenting Best Practices and Success Stories: Case studies of organizations that have successfully aligned data stewardship with sustainability goals will provide valuable lessons and inspiration for others.

(Thanks to William Hoffman for input on an earlier draft)

Back to the Blog