Making Sense within Citizen Science: Meanings and Usability of Collective Water Monitoring Data

Lake-Sea Wiki

Insufficient data on water bodies has been identified as a major obstacle to the achievement of the United Nations Sustainable Development Goals (UN SDGs) concerning water quality on a global level (United Nations 2022). In Finland, water monitoring has a long-established tradition in which citizen-collected observations, such as water levels, ice thickness, and snow depth, have played an integral role (Kuusisto 2008). The digital CS platform examined herein, the Finnish Lake-Sea Wiki, was launched in 2011 in response to volunteer citizens using various methods to submit observations to a research institute responsible for water monitoring in Finland. It offers basic information on every Finnish lake (> 1.0 ha), as well as coastal sea areas, integrating citizen observations and official data from authorities. The primary function of Wiki-based platforms is to serve as interactive knowledge infrastructures that are open to contributions from a wide community of users. Their design enables continuous updating and revision, making them self-correcting systems where collective input and peer moderation help ensure accuracy, relevance, and inclusivity over time. Accordingly, Lake-Sea Wiki allows anyone to maintain their own observation site and submit observation data. Key characteristics of the Wiki include user-driven data input and publishing, with no presumption of expert verification. Users can contribute observations, for example of blue-green algae or surface water temperature, upload photographs, engage in discussions, and use the data that others have provided. Lake-Sea Wiki has four different user groups: basic user, experienced user, expert user, and official. Expert users have received training in observations or are otherwise acknowledged as experts in water observation, whereas officials are able to add institutional monitoring data to the Wiki. In 2024, there were a total of 464,000 active users and 740,000 opened sessions in the platform.

Most CS initiatives are either developed in response to institutional needs of researchers or public authorities, or alternatively, emerge from grassroot citizen efforts typically focused on highly localised environmental concerns. Lake-Sea Wiki combines aspects of both: initiated by citizens, yet maintained by a national research institute, the Finnish Environment Institute (Syke). Syke has a central role in the monitoring of waters in Finland. It is responsible for coordinating and implementing national monitoring that generates data on the status and temporal changes of water bodies to inform policy and management. It also develops new methods to improve water quality, and promotes the sustainable use of water resources. By combining data from authorities, citizens trained by authorities, and lay-people, and transitioning to a joint digital Wiki format, Syke aimed to make this data more accessible and encourage broader community participation in environmental monitoring, raising awareness and promoting protection (Kettunen et al. 2016). The Wiki platform is cost-efficient to maintain (Kettunen et al. 2016) and benefits from a self-correcting dynamic as data accumulates through broader participation. Compared with individual CS projects, platforms tend to standardise and automate many processes that were once bespoke to projects (Baudry et al. 2021). Previously, the platform has been studied with a focus on voluntary observer motivation (Palacin et al. 2020).

Citizen science as a form of knowledge co-production from a critical data studies perspective

The concept of knowledge co-production provides a foundational lens for understanding the intertwined production of knowledge and social order; in other words, ways of knowing the world and seeking to categorise, organise, and change it (Jasanoff 2010; Stone 2016; Norström et al. 2020). In this view, digital environmental CS initiatives are not simply technical platforms for data collection but spaces where civic identities are shaped, sustainability governance is contested, and diverse groups participate, drawing on their respective positions and perspectives (Strasser et al. 2019). The practises of data production and use become embedded within broader social processes, influencing how actors perceive meanings and the actionability of knowledge (Tengö et al. 2021; Skarlatidou et al. 2024).

At the same time, CS is typically characterised as a contributory model of inclusiveness where the focus is on crowdsourcing additional data and citizens are involved solely in data collection, not present or unacknowledged in other phases of the data cycle (e.g., Franzoni et al. 2022; Land-Zandstra et al. 2021; Haklay 2013; Mahr and Dickel 2019; Peltola and Ratamäki 2024). This structure is maintained, or even reinforced, by the prevailing motivation among researchers to use CS to cost-effectively expand the spatial and temporal scope of expert observations (Frigerio et al. 2021). According to the knowledge co-production literature, effective co-production is characterised by contextual grounding, recognition of diverse ways of knowing and doing, clearly defined and shared goals, and opportunities for continuous learning, active engagement, and regular interaction among participants (Norström et al. 2020).

Critical data studies examines ostensibly neutral, depoliticised forms of data-intensive science (Iliadis and Russo 2016), tracing how data are produced, managed, and used (Bates et al. 2016; Stone 2016). It probes more deeply into the stages through which data pass over their “life course,” often visualised as a data cycle (Figure 1), with attention to the consequences of data and the related justice perspectives (Pritchard et al. 2022). The four key components of critiques are 1) data are inherently interpretive, 2) data rely on social and contextual factors for meaning, 3) data are shaped by socio-material arrangements, and 4) data function as a medium for negotiation, communication, and action (Neff et al. 2017).

Figure 1

Within its framework, as a form of knowledge co-production, CS stands among approaches that tend to overlook the processes of sensemaking in relation to the epistemological frameworks, assumptions, rationales, and socio-material complexities related to data (Hepp et al. 2022). Furthermore, compared with many other types of data-intensive science, CS adds additional layers to the process of sensemaking. The sensemaking in CS is not carried out solely by knowledge producers and seemingly detached knowledge users, but the positions of different actors are multiple and diverse (Eitzel et al. 2017; Bieszczad et al. 2023). From a critical data studies perspective, CS encompasses collective dimensions that go beyond the widely used typologies of citizen participation (e.g. Shirk et al. 2012).

The process of sensemaking in citizen science

Sensemaking is a collective, iterative process through which actors interpret and ascribe meaning to data, their production, and use (Neff et al. 2017; Stone 2016). It incorporates all actors connected to the data in different ways, both producers and users. In CS, sensemaking is not simply individual interpretation, but emerges through interactions with other individuals, observed environments, and organisational structures or platforms. This creates a need to reconcile participants’ differing standards of evidence, credibility, and value.

Embedded in the data practises, deliberate decisions, choices, negotiation, and storytelling take place throughout the phases of a data lifecycle and around socio-material arrangements: first, how data come to be; second, what kinds of meanings the data and their production have and how they get their meaning; and third, what are the proper uses of data and how are they legitimised (Stone 2016; Bates et al. 2016). For example, the objectives set for water monitoring, and the selection of equipment, platforms, observation sites, participant roles, and other socio-material arrangements have developed over a long period of time, shaping contexts of current data practises (Kuusisto 2008). These contexts are reflected in monitoring experts’ views on how monitoring could and should be conducted. The interplay between contexts and views exhibits path dependencies, partly because continuity in monitoring practises over time is itself valued. At the same time, CS has a long-standing tradition in water monitoring, and knowledge of related data practises has accumulated over time (Vuori and Korjonen-Kuusipuro 2018). Information systems formulated within CS have the potential to act as boundary objects, fostering interactive elements in collective sensemaking, translating knowledge, and supporting creation of mutual understanding (Star and Griesemer 1989).

The role of digital citizen science platforms in social processes

Through a cross-cutting process of digitalisation, environmental observations first evolved from visually perceived and notebook-recorded data points to those measured with technical instruments and stored in digital form. Now the evolution continues as digital CS increasingly adopts new functionalities and tools for recording, compiling, analysing, and communicating observations (Cieslik et al. 2018; Franzoni et al. 2022). Many CS projects are at the forefront of using new digital infrastructures and methods of environmental monitoring to facilitate data generation among the public (Franzoni et al. 2022). As the presence of digital technologies in CS is growing, they are poised to play a central part in shaping data practises (Heaton 2024; Charvolin 2024). Digital platforms shape the roles, positions, and relationships among research participants by delineating what they can and are supposed to do, what kinds of interactions are possible, what is considered relevant and valid data, what kinds of meanings the data carries, and the permissible uses of that data (Peltola and Ratamäki 2024). This represents a significant reconfiguration, creating tensions in the process, for example, around whether more established ways of observing the environment may be undermined by reliance on digital crowdsourcing (Giardullo 2023; Kennedy et al. 2022; Johnson et al. 2021).

CS platforms, seeking to stabilise participation, setting norms, and guiding behaviour, play an initial role in the process of crowds becoming a community (Baudry et al. 2021). Digitally mediated information systems, when functioning as boundary objects, can link distant communities, foster collaboration, and integrate local and scientific knowledge (Kanwal et al. 2019; Caccamo et al. 2023; Lauser 2024). Socio-material arrangements shape the specific contexts in which these information systems operate, as structured by the data cycle. In this case study, the arrangements encompass the properties of the Wiki platform, the available physical and digital equipment, the roles of the actors involved in CS and their interactions, the prevailing knowledge systems, the values and objectives of the participants, the organisational practises of their communities, and the time- and place-specific circumstances. The process of sense-making takes place throughout (Figure 2).

Figure 2

The information system of Lake-Sea Wiki as a boundary object

Compared with many of the research approaches that have been examined in critical data studies, CS involves a broad group of data producers participating in the process of sensemaking (Hepp et al. 2022). In the context of Lake-Sea Wiki, approximately 200 voluntary users have actively contributed observations during summer months, with nearly 50 continuing during winter. Furthermore, in wiki-formed CS, the roles of actors connected to the data do not divide unambiguously into producers and users of data. Here, the potential users include, in addition to volunteering citizens and water monitoring authorities providing data, other authorities, the scientific community, local communities, media, and the broader public. When examining the ability of a CS related information system to act as a boundary object, it is therefore essential to recognise this broad range of actors, their potentially shifting roles, and the differing perspectives that exist even within actor groups. For example, monitoring experts are not a homogenous group; divergent ideas about data, their production, and use emerge even among professionals (Neff et al. 2017).

In comparison to many critical data studies applications (Hepp et al. 2022), CS incorporates at least some structured elements of co-production (Norström et al. 2020), for example, by providing a platform for stabilising participation, setting norms, guiding behaviour, and exchanging different types of knowledge (Baudry et al. 2021). Here, the digital platform Lake-Sea Wiki facilitates interaction among actors in an immediate and automated manner but also, to a large extent, indirectly and unintentionally. Information flows through the platform are primarily designed to go one way, without responses or dialogic elements, even though CS inherently introduces two-way dynamics of sensemaking, for example, through the social sphere created by recorded and viewed data points, particularly when actors have multiple overlapping roles in relation to data. The recorded data points communicate what has been meaningful to observers and what they consider relevant for potential knowledge users. In turn, users, from their own perspectives and through diverse modes of use, attribute meanings to the data that were sometimes not foreseen by the producers and that may shape the producers’ future decisions.

However, the sensemaking process is collective primarily in the sense that a wide range of actors participate in it with their own perspectives. These perspectives are not necessarily shared, and their capacity to generate added value for one another is limited. Consequently, as discussed in more detail below, the actors’ interpretations of the meanings and appropriate uses of the data vary considerably. Rather than developing joint understanding, the actors gather and interpret a shared mass of data points through their own perspectives. Nonetheless, on specific occasions, as identified in the workshop by some monitoring experts and active Wiki contributors, there’s a potential to develop shared understanding. For instance, the platform enables rapid communication of acute local environmental changes, potentially allowing quick responses to emerging issues. Sometimes, a monitoring expert stated, it can also help authorities respond to emerging public concern over misunderstood but non-harmful natural phenomena, thus preventing the spread of misinformation and public anxiety.

Tensions of sensemaking: values, data quality, and actionability

CS in the context of Lake-Sea Wiki is a locally grounded activity, closely tied to environmental factors that hold local significance, and with observations primarily recorded near personally meaningful locations, such as vacation areas (58%) or homes (35%). The active Wiki contributors often valued participation for its anticipated benefits for water research, decision-making, and local water users. Some also appreciated it simply as a social hobby, and as a way of recording one’s own observations about the environment. “The good status of waters is the overall goal of the data gathering,” a Wiki contributor participating in the workshop stated, hoping that the data they produce would find use in decision-making that affects the environment. Some Wiki contributors raised concerns that actors with more political power sometimes communicate about the state of local environments in an incomplete manner and called for conveying environmental changes “reliably and without political bias.” They seek to strengthen their arguments, pursue justice through the data (Pritchard et al. 2022), and hope that the environmental changes they find meaningful will also be recognised and validated by others.

Similar to the active Wiki contributors, the monitoring experts also placed significant value on CS and its ability to contribute research, but primarily in an instrumental manner. While citizens may perceive their role in the data cycle as highly significant and wide-ranging, experts tend to assign volunteers a rather clearly defined position (see also Bieszczad et al. 2023). Expert survey answers and workshop comments often stated that the goal of CS is to enable more frequent and comprehensive data collection to complement and calibrate official monitoring efforts cost-effectively. Several monitoring experts underlined that citizen observations bring additional value particularly in areas where environmental authorities or other official bodies do not conduct regular monitoring.

However, while most of the monitoring experts explicitly stated that official monitoring cannot be replaced by citizen-generated data, they generally assessed citizen observations as commensurable with official data. In contrast to the environmental-political concerns of the citizen observers, the experts expressed concerns over the ability of CS to meet scientific quality criteria for it to be used in a complementary manner, like expert data. This perspective implies that a single citizen-generated data point should correspond to a single expert-collected data point to legitimise use. The value of citizen observations was often framed as conditional: “after all, the quality of the data is crucial – if it doesn’t meet the criteria for further use, citizen observations benefit society only in terms of engaging citizens.” When citizen observations were perceived to differ in nature and to have a higher potential for error, it was primarily framed as a quality concern that limits or often even prevents their use. An expert’s survey response stated, “it remains unclear not only what extent the quality of CS data is sufficient, but also what its appropriate uses are.”

Trust not only enables knowledge to become actionable but also defines whose knowledge counts (Kasperowski and Hagen 2022). For example, trust connects local, particular observations to aggregated data and more universal frameworks such as environmental law. The impact of CS remains limited if its contributions to the production of epistemic resources are not acknowledged (Ottinger 2022). The monitoring experts were concerned that sceptical or dismissive attitudes of authorities and decision-makers towards CS may lead to the underutilisation of citizen observations. Wiki-based CS challenges established protocols and validation procedures in which human actors interact with abstract systems to test, establish, or lose trust; however, in some of its forms, Wiki-based CS does not provide sufficient basis for trust grounded in interpersonal interaction (Kasperowski and Hagen 2022).

The approaches to standardisation and scientific quality assurance in CS, as well as its integration into sustainability assessment frameworks, remain subject to ongoing scientific debate (e.g., Fraisl et al. 2020; Balázs et al. 2021). In the workshop, the key factors enabling or restricting data utilisation were closely linked to data quality, including sufficient data volume, continuity, and consistency in observation methods and practises. However, in the analysed media discussion, the quality and comparability of official monitoring data and citizen-generated observations were not considered an issue, as the comment of a regional authority (Yle 17 June 2011) exemplifies: “Within the official monitoring system, we have a network of only 17 observation sites for algal monitoring, which covers only a small fraction of the lakes in this region. All information provided by citizens enhances our overall understanding of the state of the lakes”. This gives emphasis to the notion that the development of standards calls for insights emerging from across diverse transdisciplinary research settings that include also the potential data users (Norström et al. 2020).

Establishing unambiguous standards is not necessarily straightforward: An expert in the workshop noted that, based on comparisons, visual algae observations have been “surprisingly” consistent between experts and citizen scientists, given that there were variations in assessments among experts. Furthermore, the joint workshop revealed that citizen observers, especially those with extensive experience, have developed a strong understanding of the practical factors that influence data quality. Both the monitoring experts and active Wiki contributors had a shared interest in improving the quality of CS data and ensuring their usability. The accumulated practice-based knowledge of the Wiki contributors is currently not systematically utilised in the planning and development of observation activities and data use.

Perceived suitability of citizen science topics

The topics the active Wiki contributors reported monitoring were typically linked to specific environmental concerns, especially the visible or politically contested issues related to recreational use of waters. These topics were associated with land-use change and resulting emissions, other types of pollution, changes in water flow, impacts of climate change, changes in species occurrence, and the effects of restoration. The most observed topics were blue-green algae, surface water temperature, water transparency, and litter levels. New observations proposed by contributors included animal and vegetation species relevant for recreation or policy, and new indicators of pollution entering water bodies.

The monitoring experts, in turn, evaluated topic suitability based on measurability, simplicity, and low equipment needs. Topics requiring complex measuring equipment or extensive expertise were considered especially challenging for citizen observations. Another reason for suitability being challenged was fluctuating public interest. For example, in species monitoring, they seem to attract attention when they first appear, but tracking longer-term population development is difficult as public interest tends to decline once a species becomes abundant. From the experts’ perspective, the most suitable topics were waterbody freezing and ice-out dates as well as shoreline littering, followed by surface water temperature and unusual weather conditions.

This divergence reflects differing priorities: experts focused on data quality and methodological rigour, while citizen observers emphasised perceived relevance. Despite divided views on topic suitability, citizen contributions are already a substantial part of Finnish environmental monitoring for some topics. For example, nearly half of all blue-green algae observations are currently provided by citizens, and over 90% of citizen survey respondents had contributed algae data. Blue-green algae has also been found to be a societally relevant topic as over 80% of the analysed media articles utilised blue-green algae data. However, blue-green algae suitability did not receive unanimous support from experts: 20% of expert respondents considered it unsuitable topic for CS. Some experts found it problematic that citizens tend to record fewer “no algae” observations than experts and were concerned with drawing spatially broad conclusions from limited algae data. Nonetheless, the media often deemed it relevant to report whether citizen observations of the algae situation had been made, and if so, in what quantity, interpreting those observations as an indication of algae growth trends. Media also often used observations from just a few lakes as indicative data when publishing news reporting the regional algae situation.

Restructuring citizen science data practises for enhanced usability

The workshop discussion and the survey responses elicited development ideas focused on increasing the number of observations, fostering observer commitment, improving data quality, and ultimately, enhancing data usability. The need to expand the volume of observations was linked to debates on data quality in two ways. First, a sufficient volume of data was considered essential to ensure data reliability. Group-based participation, encouraged by hobby communities or associations, was seen as especially effective at increasing data volume and geographic coverage. To address the lack of accessible equipment, it was proposed that measurement tools could be distributed through schools, hobby groups, and public libraries. Second, expert-led quality assurance was seen as increasingly challenging as the number of observers and observations grows. The core purpose of the Wiki-formed platforms is to function as community-driven, self-correcting systems. However, the experts didn’t quite trust this model and instead called for external quality control mechanisms. Automation, through machine learning, image analysis, and visualization tools, was identified as a way to reduce manual effort. The development proposals related to increasing data volume as a means to address quality issues appear to challenge the assumption that CS data must meet the same standards as expert-produced data.

A large part of the discussion related to development proposals was connected to different forms of interaction. From the experts’ perspective, improved communication regarding data needs, targeted observation campaigns, and more regular feedback were seen as essential for sustaining citizen engagement in data gathering. Many emphasised organising observer training in partnership with non-governmental organisations, and shifting the training responsibilities gradually to these organisations. In general, the monitoring experts viewed measures that promote community engagement more positively the less resources these measures required from them: “If experts should make a significant effort and use a lot of resources for CS data, it would undermine the idea that citizen observations are an easy and inexpensive way to obtain observations”. Experts expressed concerns about managing large volumes of two-way communication, which is currently not a structured part of CS workflows. While crowdsourced big data was considered manageable, crowdsourced communication could “overwhelm,” as an expert described.

While the experts remained cautious about involving citizens in the planning and interpretation phases of the data cycle, active Wiki contributors emphasised the importance of early involvement in designing data practises, including co-developing parameter definition and input forms. One of the key messages of the Wiki contributors was that issues and development needs related to the quality of CS data should be addressed through collaboration with experienced citizen observers. The Wiki contributors called for strengthening the knowledge exchange between all actors involved in CS, including data users. They requested more active engagement that goes beyond the digital interface, feedback from experts on how data have been utilised, and co-development to improve data practises to enhance data usability.