Geography and Environmental Governance I: Governing with Environmental Data

Abstract

New data technologies abound in environmental politics. It is promised that these technologies will make the task of managing human impacts on the environment faster, easier, and more rational. Geographers have been among those eyeing this turn critically, illuminating how new data technologies are performing new environmental entities and scales of equivalence that become subject to political attention and dispute. The pulling of data through centralised infrastructure then adds layers of social judgement, as data are cleaned, formatted, and categorised to support the aims of the infrastructure's owners and managers. Throughout this process, the political economy of measurement – the idea that more measurement is a necessary precondition for more effective environmental decision making – acquires hegemony as political actors organise themselves around the politics of data and its control. With the environmental data turn reorganising the bases of environmental governance, geographical work is illuminating what is at stake and how these shifts can be engaged politically.

Introduction

With human action degrading Earth's life supporting capacity at intensifying rates, the need to reduce human impacts on the environment is stronger than ever (Ripple et al. 2017, IPBES 2024). At the same time, societal changes are driving new forms of knowledge, coordination, and ways of organising ourselves around environmental problems.

The field of environmental governance is concerned with understanding how the regulation of environmentally-impactful human actions is organised, how it is changing, and what outcomes have resulted (Lemos and Agrawal 2006, Chaffin et al. 2016, Benson and Jordan 2017, Agrawal et al. 2022, Evans and Thomas 2024). Since the 1980s scholars have observed a broad shift from national-scale command-and-control regulation of economic activities toward ‘scaling down’ of responsibility for governance outcomes to local levels, a ‘scaling up’ of authority towards transnational networks and global governance, and a ‘scaling out’ to include more private and non-state actors in environmental decision making (Reed and Bruyneel 2010, Cohen and McCarthy 2015). The essential and unique role of Indigenous People in environmental governance has also increasingly been a focus of scholarship and practice (Brondizio and Le Tourneau 2016, Thornton and Bhagwat 2021)

Amidst this changing landscape of environmental governance, geographers have made incisive contributions. Geographers have helped grasp the governance of hazards and vulnerability (Watts 1983, Burton et al. 1993, Cutter 2003), political discourses of nature (Smith 2010, Robbins 2020), the changing role of the state (McCarthy and Prudham 2004, Castree 2008, Bakker 2010), the re-scaling environmental of regulation and politics (Liverman 2004, Bulkeley 2005, Cohen and McCarthy 2015), the reorganisation of democratic participation (Bulkeley and Mol 2003, Swyngedouw 2009), the politics of environmental science (Forsyth 2003, Lave 2012, Robertson 2012), the need to – and ways to – engage Indigenous environmental politics (Coombes et al. 2012, Whyte 2017, Parsons and Fisher 2020), the embodied and emotional experiences of governance (Sultana 2015, González-Hidalgo et al. 2020), and taking seriously our accountabilities to the more-than-human world (Wright et al. 2012, Kenney-Lazar et al. 2023). Geographical scholarship continues to push boundaries by drawing attention to conceptual, empirical, and normative elements that merit incorporation into the frame of environmental governance.

With this trio of Progress Reports I spotlight three exciting frontiers of environmental governance scholarship that I believe manifest the enduring and evolving interests of geographers. This first report looks at governing with environmental data, the second at governing post-natural environments, and the third at Indigenous engagements with environmental governance. Naming these as geographical frontiers identifies the topics as distinct and valuable as well as resonating in a meaningful way with the intellectual tradition of geography. I come to these topics as a geographer who has often examined how the evolution of scientific knowledge, including data, shapes how we see and act on the environment. Although I provide a limited and highly personal take on the field, I hope these Reports can help with valuing and sustaining geographical imagination in an interdisciplinary world.

This first report reviews geographical engagements with environmental data technologies as means through which governance of humanity's impacts on the environment can be achieved. Over the past two decades, increasing calls have been made to invest in technologies to generate, organise, analyse, and use environmental data to better manage human impacts on the environment. Proponents call for investment into new observational technologies such as remote sensing, environmental DNA, drones, and GIS to know and govern the environment at large spatial scales, while new and potentially automated data sources come online for potential use in environmental decision making (Death 2015, Schenekar 2023, Schirpke et al. 2023, Anderson et al. 2025). In addition to observational technologies, one also needs to store, analyse, and then use the data – requiring investment into data warehousing infrastructure, analytical packages and methods (including artificial intelligence), and access portals and visualisation (Li et al. 2021, Güntsch et al. 2025). Across many registers, this environmental data turn is driving changes relating to how the environment is framed, how knowledge is created, and how rights and standing are accorded.

The environmental data turn has been subject to insightful analysis and critique. While mainstream boosterist perspectives emphasise the potential benefits of employing various observational, infrastructural, and analytical technologies, critical perspectives highlight how issues of automation, measurement reductionism, and public versus private ownership and control of data are reworking the political foundations of environmental governance (Asokan et al. 2020, de Albuquerque et al. 2021, Kloppenburg et al. 2022). Amongst these debates, geographers and their collaborators have explored these issues in thoughtful ways, including recent collections by Goldstein and Nost (2022a), Nost and Goldstein (2021), Loconto et al. (2024) and Molle et al. (2024). With this Progress Report I take stock of this wave of interest by asking: how is geographical scholarship helping to understand efforts to govern with environmental data?

Performativity of Measurement

Critical studies of environmental data elaborate how the process of making data through measurement simplifies what is represented and also creates something new that can be acted upon (Loconto et al. 2024, Molle et al. 2024). Measuring a coral reef by remote sensing or estimating water extraction through quantitative calculations creates a data-fied representation of the system in which some ecological features are accounted for, others are not, and one environment can be compared to another along a single quantitative axis – a process called commensuration (Espeland and Stevens 1998). Geographers, sociologists, and anthropologists have over three decades contributed robust empirical analyses and conceptual critiques of efforts to ‘make environments the same’ through technical forms of measurement and analysis (Rocheleau 1995, Robbins 2001, Robertson 2006, 2012, Höhler and Ziegler 2010, Turnhout et al. 2014, Mennicken and Espeland 2019, Cusworth and Stanley 2025). This datafication of the environment through measurement is a process of “dematerialization that converts natural phenomena into symbolic material that can be indexed and searched” (Wickberg et al. 2024: 2).

Several recent analyses carry forward the concern with commensuration and performativity of measurement into global discourses (Wickberg et al. 2024). Linton and Saade (2024) show how hegemonic framings of ‘global water’ perform water as a commensurable quantum and that this abstraction of water out of its social relations misses the opportunity to discuss what those social relations should be instead. Puy and Lankford (2024) critically analyse the claim that the globe is close to breaching a ‘planetary boundary’ for water, finding that the claim's numerical foundations and socio-economic assumptions lack justification while at the same time narrow the suite of political options. In the wider domain of the global environment and development, scholars illuminate how global indicators for measuring progress are performative. A systematic review by Hagerman et al. (2021) found that formal policy indicators reflect only those parts of the societal goals that are most easily measured, rather than the most valued or important, and that indicators tend to reflect the long-standing institutional commitments of the participating organisations. In relation to the Sustainable Development Goals, MDee et al. (2024: 410) concluded that the indicators for SDGA 6.5.1 ‘are subjective, contested, and largely meaningless’.

Measurement is prized for its purported objectivity but underneath it is infused with values (Pine and Liboiron 2015, Loconto et al. 2024). The perceived objectivity of measurement comes from its claim to directly reference the material world and the fact that its methods and units are standardised and therefore open to scrutiny and validation. However, values inform which paths to measurement should be taken and what to measure in the first place (Blue and Brierley 2016, Elliot 2017, King and Tadaki 2018). Nost (2024) shows, for example, how models of sediment diversion in Louisiana intended to guide wetland restoration can produce drastically different results depending on which values-based assumptions underpin the analysis (see also Braun 2021 for an example from remote sensing). Molle and Collard (2024) show how environmental flows – managing water flow in a river to support a range of values – was intended to provide objective voice to articulate the ecological needs of a river, yet the assumptions of what is needed for a river are a shaped by current political appetite for regulation of industry. Acharya (2024) reveals how the measurement of ‘total dissolved solids’ in Indian drinking water is putatively scientific on the one hand, yet embeds contestable judgements about the commensurability of chemical sources on the other, and this commensuration carries significant consequences for state investment and marketisation of water. Bhatt (2024) clarifies how comparative measurement effectively provides the goalposts for environmental management, rendering all that is not measured as less visible and hence less amenable to political prioritisation.

Frontiers in environmental measurement are also reconfiguring space and spatial politics (Garrett and Anderson 2018, Anderson et al. 2025). Millner et al. (2024) elaborate how drones provide a vertical and volumetric perspective, making it straightforward to see humans entering and leaving a protected forest, for example. However, without seeing what people have done inside the forest, this leaves open ambiguity and the opportunity for cultural bias and stereotyping to fill in the gaps, with punitive consequences (Millner et al. 2024). Ritts et al. (2024) show how emerging datascapes of environmental sound are similarly reorganising disciplinary power. Acoustic monitoring undertaken for ecological data gathering, such as to observe birds in a given location, can also register gunshots (assumed to be from poachers), which can help direct policing resources in space. This privileging of automatically-detectable soundwaves, however, leads to reorganising scarce resources around this data signal, prioritising the logic of distinct sonic detectability over other evidence and other reasons for distributing enforcement resources. Environmental measurement performs space selectively, and the actors and interests invested in these technologies are actively encouraging reorganisation of society around these selections.

Infrastructures as Centralised and Embedded Judgement

Once generated, data travel through digital infrastructures for cleaning and standardisation, analysis, and use.

As defined by Slota and Bowker (2017: 537) building on Star and Ruhleder (1996):

Infrastructure… is embedded and transparent; infrastructure exists (metaphorically) within or underneath other social, technological, and built worlds and does not need to be reconsidered at the moment of each task it enables. Infrastructure is learned as a part of membership in a given community and linked with the conventions of practice therein and embodies some set of standards. It is built above an installed base, becomes visible upon breakdown, and is of a scale or scope that exceeds a single “site”—however that might be conceived.

Environmental data infrastructures such as server farms, data management technologies, and data access portals shape what is conceivable and considered doable in environmental governance with data. Such infrastructures consolidate social relationships and can deepen asymmetries, they mediate material flows as well as transform symbolic ideas about the environment, and they are shaped by – and in turn shape – particular places (Goldstein and Nost 2022b).

Despite images of ‘the cloud’ as sitting above place-relations, data infrastructures are materially placed (Nost and Goldstein 2021). Data centres for example require a complex socioeconomic metabolism involving e-waste and heat waste, land ownership, electricity and water use (Ritts 2026). Pickren (2018) argues that the geographies of data centres are propelled into particular directions by the costs of land, materials and related infrastructure such as energy and law, and the organisation of labour. These geographies in turn generate ongoing effects as their location subjects them to certain laws and relational influences, they consume certain local resources, and they create proximal markets (e.g., housing, retail) and political constituencies (Lally et al. 2022, Nost and Colven 2022, Ritts 2026).

As data pass through a common infrastructure for cleaning and standardisation, value-laden judgements are applied. Data cleaning is one such process of centralised quality control in which criteria are applied to include, exclude, and modify data. Clifford (2022) describes how spikes in dust levels that exceed air quality standards in the United States are often creatively coded by government bureaucrats as ‘exceptional events’ that therefore do not merit status as a regulatory breach. Similarly, Berry et al. (2024) show how the Nevada Division of Water Resources in the US categorise data on water use, applying creative judgement to classify quanta of mining water use as ‘temporary’ and ‘non-consumptive’, which excludes them from regulatory categories, calculations, and controls. How governments and scientists include or exclude extreme or ‘outlier’ data in climate and atmospheric science fundamentally shapes the baseline understanding of the system, and therefore what changes merit public concern (Clifford and Travis 2021).

Alongside the ‘cleaning’ of data, data are also transformed through formatting to make them legible to different systems of analysis. Ghosh (2024), for example, shows that government administrators prefer data formats that align with existing data systems, whereas scientific experts prefer data formats that align with other scientific models of understanding. Data may need to be consistently formatted for potential analysis by artificial intelligence, for example, which is yet another centralised application of judgement to data through infrastructure (Maniyar et al. 2025).

Together, the centralisation of data through infrastructure constitutes a key mechanism of knowledge politics. The bringing of data together through a shared infrastructure subjects them to transformation through application of judgement – about what data to include/exclude, and what preferences for categories and formatting should take precedence. As Mennicken and Salais (2022: 26) summarise:

Rather than being rooted in collective deliberation, the informational basis of governance-driven quantification is predetermined and imposed by “the Centre”; it incorporates norms without discussion and directs the decisional process towards prefixed political outcomes.

As infrastructure centralises control over data, the question of who owns the infrastructure – whose judgement and goals direct its use – becomes a structural feature of environmental politics (Nost and Goldstein 2021). Who controls who can access data, use it, and under what conditions? Underneath public narratives of transparency and open data lie specific infrastructures like data portals, analytical procedures and output visualisations that are controlled by the judgement of specific actors ranging from state agencies, contractors, and private corporations, to non-government organisations, Indigenous and community groups (Goldstein and Faxon 2022, Kloppenburg et al. 2022, Lin 2022). The Indigenous Data Sovereignty movement insists that Indigenous communities should control data that is generated with, by, and for their communities (Kukutai and Taylor 2016, Walter et al. 2020). This is because Western science and research practices have historically dispossessed Indigenous people of their knowledge by extracting it, interpreting it, and wielding it without their permission, guidance, influence, or benefit (Tuhiwai Smith 2021). In an agricultural setting, Fairbairn and Kish (2022) show how data brokers in the Global South work to convince smallholders to share their data on the promise of benefit but without subsequent free access to the knowledge products that the brokers produce, which require hired expertise to use. The promise of data-driven environmental governance funnels the power to represent people and the environment into the hands of a small number of infrastructure owners, who may pursue goals that include but also exceed public good or the specific benefit of participants. Goldstein and Faxon (2022) show how purportedly ‘transparent’ infrastructures developed for democratising spatial data on forests in Myanmar are in practice constrained through legal as well as political forces to a tight circle of government agencies and networks: “Satellite data might provide tools to refute the government, but today those tools remain predominantly in the hands of foreign organizations and a few urban Myanmar elite” (p53).

In these ways, struggles over environmental data infrastructures are effectively struggles over the ability to make authoritative claims about environmental change that are underpinned by one's own values and objectives. As such, critical scholars have argued that community participation in and ownership of data infrastructures is key, advocating for practices including: Global South nations launching and owning their own satellites and remote sensing infrastructures (Alvarez León 2022), communities generating data that have ‘friction’ or misalignment with state infrastructures to enable and sustain local control over data (Johnson et al. 2022), remote sensing scholars aligning with community goals and governance (Segarra et al. 2024), and communities generating data for themselves rather than letting the state's ignorance continue to provide a basis for inaction on environmental pollution (Gabrys et al. 2016).

Political Economy of Environmental Observation

Investments into environmental data technologies are uneven and are changing, driven by multiple forces. A largescale shift has been occurring for some time from intensive field-based methods and technologies towards remotely-sensed and automated technologies. Coral reef ecosystems are being known less by dive surveys and more by remote sensing, buoy systems, and modelling (Braverman 2022). For forests, in-situ surveys and interactional knowledge of forests are being devalued as investment is funnelled instead toward remote sensing applications that enable claims about increasing or decreasing forest extent (Gabrys et al. 2022). In the wider ocean sciences, this means shedding reliance on capital-intensive observational technologies such as submersibles carried on research vessels and toward modelling ocean dynamics at a distance (Lehman 2018). Physical geographers are among those lamenting the decline of field-based methods such site-based landform interpretation (Fryirs et al. 2019), whose intensive information about place-based physical processes can explain landscape behaviour better than spatially-extensive remote sensing or proxy observations, however numerous (Houser et al. 2022, Braun 2024, see also Salmond et al. 2017).

Shifts in environmental measurement are driven by a mix of scientific, public, and private motivations that can overlap. Scientists emphasise how new data technologies unlock new understanding through their expanded physical reach, resolution, speed, or accessibility (Guo et al. 2015, Stephens et al. 2020, Shen et al. 2024, Anderson et al. 2025). Public and quasi-public actors collect data and promote data infrastructures that can help hold states and private actors to account (Kloppenburg et al. 2022, Tadaki 2024). Private actors develop ways to generate and assemble social and environmental data into saleable commodities that return a profit (Robertson 2012, Zuboff 2019, de Albuquerque et al. 2021). These motivations are not exclusive but interact of course, as “science produces the raw materials for subsequent control and exchange” (Turnhout et al. 2014: 581) and the infrastructures and data used by public agencies may even be owned by private actors in the first place. The co-production of environmental data through the logics of scientific understanding, public policy, and private profit thus iteratively shapes what is known and can be known about the environment.

A key rationale for ongoing investment into environmental measurement is that rigorous, standardised, scientifically authoritative data provide the currency for policy and regulatory decision making. Measurements and data that are considered scientifically rigorous are prized and accorded weight in technocratic decision making processes such as standard-setting or the courts. Environmental DNA is marketed as prestigious due to its highly-technical discourse, yet interpretive challenges constrain its reliability (Shen et al. 2024). Loring et al. (2021) argue that obtaining moral standing in environmental governance is in practice achieved more from one's ability to produce and wield authoritative ‘data’ than simply emanating from one's inherent dignity as a citizen. In postcolonial contexts such as Aotearoa New Zealand, even when the state may subsidize the creation of environmental data by Indigenous communities guided by Indigenous knowledge, the state can still undertake ‘cherry picking’ of data that is assimilable within Western science frameworks on the same topic (Tadaki et al. 2022). The need for data to be recognisable by the state also drives community groups to organise citizen science initiatives to align with state formats that will be recognised (Cohen et al. 2021, Gabrys and Prichard 2022, Hesse et al. 2023).

The idea that more data will enable better environmental protection is increasingly being contested. Findlater et al. (2021) observe that while ‘climate services’ programmes promise better decisions for governments and communities, in practice they focus on generating more sophisticated data. In the ecosystem services field, Primmer and Furman (2024) report that while studies promise that ecosystem valuation will generate better decision making outcomes, the vast majority of work simply produces more valuation data; there is little insight into whether any improved outcomes are visible. In freshwater policy, Tadaki (2024) shows how even under a strong evidence-based regulatory framework, bespoke high-resolution data can fail to justify environmental improvement. Across these and other accounts “data is seen to be the necessary precondition for understanding and acting on environmental change” (Gabrys et al. 2022: 66) and yet it is power that decides when data is considered sufficient and what decisions should be rendered. Orosco and Ybarra (2025: 13) argue that “datafication will not bring environmental justice, but accountable relationships to place will.”

Powerful actors control environmental politics through what Shapiro et al. (2017) call the data treadmill, which refers to the constant effort yet ever-receding goal of obtaining measurements that are accepted as sufficiently valid for regulating human actions on the environment. Together scientists, regulated industries, and regulators promote the promise of more measurement and the use of the data treadmill (Hesse et al. 2023). This provides a logic for continuing investment into environmental science while also deferring consequential regulatory action. But importantly, while some environmental entities become the focus for measurement, other entities continue to sit outside this focus and remain inscrutable – and thus un-regulatable – by the state (Kroepsch and Clifford 2022, Simon 2022, Vurdubakis and Rajao 2022, Kroepsch 2024). The political drivers of investment into measurement produce light but also shadow.

Conclusion

Emerging efforts to govern with environmental data are reorganising the foundations of environmental politics. New observation technologies are modifying the environmental entities to be accorded value and standing in decision making, and new infrastructures are reshaping who participates in producing, using, and benefitting from environmental knowledge (Bakker and Ritts 2018, Johns 2021, Nost and Goldstein 2021, Loconto et al. 2024, Molle et al. 2024). Both environmental observation and data infrastructures are in turn driven by a political promise that more measurement and centralised data will make governing the environment easier, yet this promise often fails to manifest in practice.

The contributions covered here carry forward longstanding themes in geography on the social construction of nature, the material logics and effects of uneven capitalist development, and environmental justice. Geographers and others are making incisive contributions to understanding these developments, in both critical and practical registers. In a critical register, this work is elaborating the ontological politics of measurement and infrastructuring environmental data, showing how the transformations proposed and wrought by environmental data technologies are reworking humanity's relationship with the material environment.

In a practical register, geographical work on environmental data technologies also move beyond critique to show how this constitutional turn might be seized to empower the people, values, and voices of those previously marginalised. Geographers alone can’t stop or create these wider technological changes, of course, but by foregrounding and the scientific and public good aspects of environmental data technologies, geographers are helping to illuminate values-based paths forward through these transformations in environmental governance.