Establishing a long‐term citizen science project? Lessons learned from the Community Lake Ice Collaboration spanning over 30 yr and 1000 lakes

Recruiting the public to participate in the scientific process can be invaluable in furthering our understanding of global environmental change. Several long‐term citizen science projects have been active for over a decade, with most involving the public in the data collection phase of the scientific process. Our team has recently inherited a long‐term citizen science project called the Community Lake Ice Collaboration (CLIC). We have benefited from engaging with a community of experienced participants who have collected reliable data for hundreds of lakes across the USA for over 30 yr. Collecting data at this scale would have been logistically and financially challenging without participants volunteering their time and effort. Here, we aim to provide an overview of the lessons we have learned from CLIC and other citizen science projects and develop nine guidelines for establishing and sustaining a long‐term citizen science project.

financially challenging without participants volunteering their time and effort.Here, we aim to provide an overview of the lessons we have learned from CLIC and other citizen science projects and develop nine guidelines for establishing and sustaining a long-term citizen science project.
Various environmental stressors currently threaten lakes (Witte et al. 2000;Hilborn et al. 2003;Schwarzenbach et al. 2006;Bernhardt et al. 2017;Woolway and Maberly 2020), with climate change being one of the most pressing and complex issues (Woolway and Maberly 2020).The response of lakes to climate change is evident via warming water temperatures (O'Reilly et al. 2015), changing mixing regimes (Woolway and Merchant 2019), earlier and longer-lasting thermal stratification (Woolway et al. 2021b), heat waves (Woolway et al. 2021a), declining oxygen levels (Jane et al. 2021), changes in lake thermal habitat (Kraemer et al. 2021), increasing salinity in freshwater systems (Roy and Zahid 2021;Melles et al. 2023), and the loss of lake ice coverage (Sharma et al. 2019(Sharma et al. , 2021)).To document these changes, scientists rely on in situ observations, highfrequency sensors, experiments, paleolimnological methods, remote sensing, and empirical and mechanistic modeling (Sharma et al. 2020).However, some of the earliest and longest direct observations of lakes have been maintained by citizen scientists (Sharma et al. 2022).For example, the lake ice phenology record for Lake Suwa in Japan extends back to 1443, collected by 15 generations of Shinto priests (Sharma et al. 2016;Knoll et al. 2019).The information produced by citizen science can significantly contribute to understanding and addressing complex and sometimes ill-defined consequences of climate change and other environmental stressors.
However, to understand and address environmental stressors at a regional or global level, it is essential to have information that covers vast spatial and temporal scales and at fine resolutions (Fritz et al. 2019).To rely solely on professional scientists to collect the required in situ observational data would be timeconsuming, expensive, and temporally and spatially limited (Dickinson et al. 2010;Franzoni and Sauermann 2014).A solution to this challenge is recruiting the public's help through citizen science (Silvertown 2009).With public participants (hereafter referred to as participants) volunteering their time and effort, it is more feasible to obtain the required data efficiently and cost-effectively (Fritz et al. 2019).The data collected from citizen science can be used for various purposes in addition to advancing scientific research, including environmental monitoring, validating remote sensing observations, influencing resource management, informing decision-makers, and educating participants (Shirk et al. 2012;Little et al. 2021).
With the benefits of citizen science becoming more recognized, a growing number of projects have been established in recent years (Silvertown 2009).Most of these projects focus on understanding and addressing the consequences of ecological and environmental change (Silvertown 2009;Cunha et al. 2017).However, there are different approaches to accomplishing these objectives.For example, the most common citizen science projects are classified as contributory.These projects involve professional project leaders (e.g., scientists practitioners; hereafter referred to as project leaders) establishing and designing the project and recruiting participants to help with data collection (Bonney et al. 2009;Shirk et al. 2012;Haklay 2013).Citizen science projects can also be collaborative or co-created, which are classified based on the degree of participant involvement (Bonney et al. 2009;Shirk et al. 2012;Haklay 2013;Jordan et al. 2016).In collaborative projects, establishment and design are mainly the responsibility of the project leaders, similar to contributory types (Bonney et al. 2009;Shirk et al. 2012;Haklay 2013).However, these projects also involve participants in other elements besides collecting data, such as refining project design, analyzing data, and communicating findings to the public (Bonney et al. 2009;Shirk et al. 2012;Haklay 2013;Jordan et al. 2016).Cocreated projects take a different approach, involving some participants in most or all project elements.For example, participants may be involved in developing goals, project design, acquiring funding, collecting and analyzing data, communicating findings, and developing future questions (Bonney et al. 2009;Shirk et al. 2012).
In addition to the three project types, an interesting subset of citizen science projects are those sustained long-term.Longterm projects benefit from continuous monitoring, conservation, or advocacy work for longer periods.Several examples of these projects exist, spanning a decade or more.These include the Audubon Christmas Bird Count, established in 1900 and recognized as the longest-running citizen science project in North America (Butcher 1990).Additional examples include Riverkeeper, established in New York in 1966 (Cronin and Kennedy 1997), the Great Whale Count, established in the United States and the United Kingdom in 1991 (Tonachella et al. 2012), and Cornell Lab's NestWatch, established in 1996(Bonney et al. 2009).An essential element of sustaining a longterm project is motivating individual participants to continue their involvement, which requires project leaders to focus time, energy, and resources on this purpose (Cooper et al. 2007).Sustaining participation is associated with unique benefits, including fewer data gaps and higher quality data from more experienced participants (e.g., Cooper et al. 2007;Cunha et al. 2017).There is also the opportunity to build a core group of experienced participants who can fulfill local leadership roles and provide support in more advanced ways (e.g., Cooper et al. 2007;Cunha et al. 2017).
Many lessons can be learned from established, long-term projects.Fortunately, we recently inherited the Community Lake Ice Collaboration (CLIC) from its founder Kenton Stewart upon his retirement.This project was established in the 1980s when long-term evidence was limited to confirm how humans affect the climate.CLIC aimed to fill this data gap by collecting lake ice-on and ice-off dates from participants via postcards across five states: Maine, Michigan, Minnesota, New York, and Wisconsin (Fig. 1).The purpose of these data was to track lake ice and climate trends over time.To date, this project has collected over 52,500 lake ice phenology observations for 1008 lakes and involved 935 monitors over more than 30 yr (Fig. 2; Sharma et al. 2023).
With the inheritance of CLIC, we have learned more about the intricacies involved in citizen science.Therefore, we have created guidelines based on the lessons learned from running CLIC, our past citizen science experiences, and consulting the published literature from interdisciplinary fields.As most citizen science projects, including CLIC, are classified as contributory, we provide guidelines that pertain to contributory projects.We aim to help new projects become established and existing projects to continue long-term.

Citizen science guidelines
Below, we provide nine guidelines for establishing and sustaining long-term citizen science projects.The guidelines are: (1) establish project goals; (2) acquire funding; (3) think about barriers to participation and how to overcome them; (4) form a leadership team and collaborate with other organizations; (5) collect simple data or allocate resources to verify data quality; (6) develop a strategy for recruiting participants; (7) monitor and manage data submitted and technology used in the project; (8) monitor and evaluate the project; and (9) engage with participants and encourage participant interaction (Fig. 3).Here, we provide examples of how we implemented each guideline in CLIC, tailored to the project's goals and needs.However, other long-term citizen science projects have also implemented most or all of these guidelines, including Nature's Notebook by The USA National Phenology Network and eBird managed by the Cornell Lab of Ornithology.
We note that the purpose of these guidelines is to serve as a starting point and to bring awareness to increasingly discussed topics, including diversity and inclusivity.
Guideline 1: Establish project goals Generally, there are two approaches to establishing project goals.The first approach is defining the goals before establishing the project (Chambers et al. 2021) since they can affect several elements, such as project design and data collection.This approach can help provide a clear idea of how to invest time, effort, and resources.
The second approach is allowing goals to develop as the project progresses.Goals can be adjusted as new, relevant opportunities emerge or as project leaders and participants share perspectives and knowledge, potentially leading to further questions and considerations (Burke and Heynen 2014).Projects can use both approaches.Goals may be established at the beginning of the project and evolve later, or new ones may be added to the original purpose.For CLIC, the project aims to track current lake ice and climate trends with the data collected.For example, the State of the Climate reports in 2022 and 2023 used a subset of CLIC's ice data (Dunn et al. 2022).We also plan to use the data in limnological studies once the quality is suitable for scientific research.However, new goals have emerged as the internet and information and communication technologies (ICTs) have become more accessible.We identified the opportunity to share data with participants via an exclusive website, displaying submitted data within a graph or table and providing the mean, maximum, and minimum ice dates on record.Participants can view the data for their assigned lake and all other lakes monitored for CLIC since the 1980s.While we continue the original trajectory of CLIC, using the data for monitoring and research purposes, we have also evolved our goals to include sharing the data with participants.Currently, we are working with  participants to update the ice records, validate the ice dates, and fill in data gaps.We will share these data publicly, likely submitting them to the Environmental Data Initiative or National Snow and Ice Data Center databases once these quality assurance procedures are completed.

Guideline 2: Acquire funding
Although citizen science is a cost-effective way to gather data and educate the public, some project elements still require funds, despite participants absorbing much of the costs by volunteering their time and effort.Expenses include participant training, equipment to collect data, data screening and checking, data organization and storage, communication with participants, and research coordination (Deluca et al. 2010;Carballo-C ardenas and Tobi 2016;Merenlender et al. 2016;Pateman et al. 2021).Often, a professional leadership team is financially compensated for completing these tasks.Therefore, it is important to acquire funding before a project begins to prevent avoidable budget issues later (Whitelaw et al. 2003).Funding can be obtained from several institutions, including universities, NGOs, and public and private agencies (Cunha et al. 2017).The ideal approach is acquiring funding from multiple sources (Conrad and Hilchey 2011) as a contingency since it provides more consistent and diverse support than relying on a single source (Latimore and Steen 2014).However, according to Cunha et al. (2017), among 126 citizen science projects between 2005 and 2014, only 32% obtained funding from multiple sources.
One of the most common challenges when establishing or sustaining a project is the limited funding sources available, especially for long-term purposes (Chase and Levine 2016;Turrini et al. 2018).We have encountered this challenge at CLIC, relying on mostly two sources to fund our citizen science activities.Both sources are also not tailored for long-term projects.The lack of long-term funding leads to more time and effort spent searching and applying for potential grants.It is also challenging to ensure a project will continue longterm when funding is only secured annually or for a few years.These challenges highlight the need for collaboration, which can provide monetary and non-monetary support, and will be further discussed in guideline 4.

Guideline 3: Think about barriers to participation and how to overcome them
Inclusivity is at the foundation of citizen science as it allows the public to participate in a process that can help shape their world personally or through political and social change (Johnson et al. 2014;Cooper et al. 2021;Pateman et al. 2021).However, currently, there is a lack of diversity among citizen science participants, with an underrepresentation of Indigenous and Black individuals and people of color.Also underrepresented are individuals with lower income, limited access to science education and scientific references, lower social capital (lack the benefits of being part of social networks) (Portes 1998), and less leisure time (Benyei et al. 2021).Consequently, these individuals are excluded from receiving the benefits of participating in citizen science and bringing forth issues that affect their communities (Purcell et al. 2012).
Lack of participant diversity may lead to misrepresentation of environmental conditions since there is often a connection between environmental variables and demographic factors (Purcell et al. 2012).Regions with higher levels of socioeconomic deprivation typically have lower environmental quality (Fairburn et al. 2009).Therefore, increasing participant diversity can lead to unbiased data collection, especially in historically undersampled areas (West and Pateman 2016), which can help better inform resource management decisions and policies for these areas.Moreover, including underrepresented groups can provide the opportunity to incorporate local and place-based knowledge into the Western scientific process (Bäckstrand 2003;Lidskog 2008).This knowledge is essential when tackling complex issues such as biodiversity loss, pollution, and climate change, where local, regional, and global information are (Bäckstrand 2003;Lidskog 2008;Bonney et al. 2014).
To overcome barriers, project leaders need to dedicate time, effort, and sometimes funds to create solutions.For example, to help eliminate time barriers, projects can have diverse tasks, with some requiring less time than others (Purcell et al. 2012).If participation costs create barriers, such as transportation or equipment (Merenlender et al. 2016), allocating funds to cover these costs can open the opportunity for underrepresented groups to participate.Similarly, inclusivity can be achieved for projects incorporating technology by allowing participants to be involved with or without using technology.For example, CLIC participants can submit lake ice data using postcards with prepaid postage or by phone, website submission, or email, which allows participation regardless of technical skills or access to technological resources.

Guideline 4: Form a leadership team and collaborate with other organizations
Collaboration and incorporating interdisciplinary knowledge have been significant in addressing environmental stressors (Odum and Barret 1971).However, with the increasing complexity of these stressors (e.g., Dawson et al. 2011), collaboration has become even more crucial.Therefore, two types of collaborations can be considered when establishing a citizen science project.First, forming a leadership team responsible for running the daily elements of the project and second, working with organizations or local groups with similar goals.
A leadership team is necessary to sustain a citizen science project because it involves many moving parts.Multiple project leaders are required to coordinate tasks and share various responsibilities to keep a project well-organized and running efficiently.Several tasks require continuous attention in a project, such as data screening and checking, providing feedback, recruiting, training, planning outreach events, coordinating program logistics, developing educational content, acquiring and managing funds, and coordinating work among other organizations (Chase and Levine 2016;de Moor et al. 2019).At CLIC, for example, even when the budget was limited without external funding, at least three people fulfilled multiple roles to sustain the project.The team was responsible for entering data, communicating with participants, and handling the logistics of sending letters and postcards to engage participants and obtain data.Currently, we have a team of five to six people with the same responsibilities and additional roles, including managing data and the project website, ensuring data quality, researching, creating a newsletter, and organizing and executing outreach activities, such as our annual conference.
The second type of collaboration involves working with other organizations, combining resources and efforts to make a new project possible or sustain an existing one.As mentioned in guideline 2, obtaining funds for a long-term project can be challenging, but forming collaborations can help overcome this challenge.For example, when the Florida Fish and Wildlife Conservation Commission received limited funding for its conservation work, the government agency sought collaborations with multiple businesses and community organizations (Dutterer et al. 2014).The TrophyCatch program, which recruits the help of anglers to report trophy bass caught, was formed from these collaborations (Dutterer et al. 2014).The Florida Fish and Wildlife Conservation Commission gained in-kind support, including assistance in handling program logistics and acquiring prizes and incentives, reducing the program's financial burden (Dutterer et al. 2014).As a result, the agency has been able to continue its conservation work for the last 10 yr.
Other benefits associated with collaborative work include sharing techniques and advice, developing new questions, and helping create consistent procedures (e.g., collecting or validating data) for working in the same ecosystem (Freitag and Pfeffer 2013) or monitoring the same resource.These exchanges can also help reduce redundancy and inefficient use of limited resources by preventing the scope of similar projects from overlapping extensively (Freitag and Pfeffer 2013;Carballo-C ardenas and Tobi 2016).Particularly for newer projects, collaborating with well-known, respected, and trusted organizations with similar goals can help motivate participants to join a project and help sustain participation over time (Benyei et al. 2021).
Guideline 5: Collect simple data or allocate resources to verify data quality Data collection is often an element of citizen science projects (Cooper and Lewenstein 2016;Turrini et al. 2018).Fortunately, these data have become increasingly accepted in scientific, resource management, and political communities due to evidence verifying the comparability of citizen science data to professionally collected data (Hoyer et al. 2014;Stepenuck and Genskow 2018).Regardless of this evidence, there is still some uncertainty in using these data for scientific studies or decision-making (Hoyer et al. 2014;Stepenuck and Genskow 2018).Approaches that can help minimize uncertainty include collecting simple data (e.g., observational) or allocating time, effort, or funds to ensure the collection of high-quality, complex data (e.g., quantitative [Dickinson et al. 2010;Follett and Strezov 2015;Carballo-C ardenas and Tobi 2016]).At CLIC, we collect simple observational data from participants: lake ice-on and ice-off dates.There is only one criterion for each event; the lake must be completely frozen or completely ice-free.Keeping the data and criterion simple eliminates the need for training while still obtaining highquality and consistent data.And although the data is simple, we acquire useful information about the lake and climate.
One of the first steps in determining the data required for a project is identifying the level of detail needed to understand the issue of interest.For example, if a project requires participants to identify organisms, determine if higher classification levels (e.g., genus or family), which are easier to identify (Bloniarz and Ryan 1996;Brandon et al. 2003), provide sufficient information for the intended use of the data.However, suppose the species level of classification is required.In this case, choosing species with distinct physical features or incorporating mobile applications that use visual recognition software such as Leafsnap or iNaturalist can increase the accuracy of identification (Crall et al. 2011;Preece 2016).Although some of these adjustments may simplify the data, valuable and accurate information can be acquired while minimizing errors and reducing the need for training.
If more complex data are required for the project's intended use, project leaders should be prepared to dedicate time, effort, and funds to ensure high-quality data are collected.Several approaches can be implemented, including training participants using a standard protocol, supervising data collection, and organizing group data collection and entry (Dickinson et al. 2010;Bird et al. 2014;Carballo-C ardenas and Tobi 2016;Swanson et al. 2016).While it is more challenging to ensure high-quality, complex data, approaches are available to overcome these challenges but require more resources.

Guideline 6: Develop a strategy for recruiting participants
A crucial part of recruiting participants is identifying what encourages the public to participate in citizen science.Therefore, the first step in a recruitment strategy is to understand issues that are of public interest.For example, it is typically easier to recruit participants for projects that focus on keystone, indicator, or charismatic species (e.g., marine mammals) and species that perform ecosystem services (e.g., pollinators [Tonachella et al. 2012;Graham et al. 2014;Chase and Levine 2016]).There is also a high interest in projects that focus on resources essential to human life (e.g., water [Canfield et al. 2002;Chase and Levine 2016]).At CLIC, most participants live or work near their assigned lakes; therefore, the interest was likely already present and encouraged the initial recruitment.However, if aiming to increase the diversity of participants, for example, in education level, race, ethnicity, or age, additional strategizing is required.Consider the needs, motivations, and interests (Pandya 2012) of typically underrepresented groups to increase diversity.In addition, aligning the project's agenda and design with these interests (Pateman et al. 2021) or providing the opportunity to codesign the project (Asingizwe et al. 2020) with these communities can help recruit diverse participants.
To recruit participants, another step to strategize is how the opportunity will be advertised.There are different approaches to advertising, including nontargeted and targeted approaches.Nontargeted advertising informs many people about the opportunity via mass media or social media (van Vliet et al. 2014;Brouwer and Hessels 2019).However, this approach tends to attract a higher number of people that are already inclined to participate in citizen science, such as highly educated individuals (Brossard et al. 2005;Crall et al. 2013).Similarly, advertising, such as through word-of-mouth, can have a similar effect.However, by slightly adjusting the targeted approach, Brouwer and Hessels (2019) demonstrated the possibility of increasing participant diversity.In this study, the authors compared targeted and nontargeted advertising in five surface and drinking water projects.Their adjustment included emailing personal invitations to a random sample of households.Compared to nontargeted approaches (e.g., Facebook campaign), this significantly increased the diversity in education level and recruited more participants of older age groups.
At CLIC, initial recruitment occurred using a targeted approach, informing family, friends, and colleagues about the opportunity.However, an adjusted targeted approach was later implemented by directly calling local chambers of commerce, marinas, and state conservation officers that lived or worked near lakes, expanding the network of participants.Currently, CLIC does not actively recruit since the project has been wellestablished for a few decades.Our main focus since inheriting CLIC has been to support participants with the changes made to the project (e.g., new project leaders and online data submissions).We have also focused on identifying participants actively submitting data, understanding other project logistics, and filling in gaps in the lake ice data records.However, we have shared the story of CLIC at conferences and interviews, which has helped recruit new participants.In the future, we will begin succession planning, recruiting individuals to replace participants who can no longer participate and potentially actively recruiting more individuals to join.
Guideline 7: Monitor and manage data submitted and the technology used in the project While many of these guidelines are revisited throughout a citizen science project, some are revisited more frequently than others (Fig. 3).The frequency depends on the project, with its different goals and needs.At CLIC, we constantly monitor and manage data and the websites we use for submitting and viewing data.Because our goal is to reach the appropriate level of data quality for research and decisionmaking, we frequently re-engage with this guideline.In addition, creating an enjoyable and stress-free experience for CLIC participants is a priority, which includes ensuring that the project's websites function correctly.
Regarding data monitoring and managing, a helpful approach is to develop a standard operating procedure (SOP) that remains relatively consistent over time.An SOP helps validate the level of data quality (e.g., for scientific studies).An excellent example of an SOP is implemented by the contributory project eBird which receives millions of bird sightings annually (Sullivan et al. 2009).Its procedure incorporates filters to flag questionable data submitted online, automatic messages asking the participant to provide more information about flagged data, and reviews by a human expert if questionable data are still submitted (Sullivan et al. 2009;Preece 2016).
Similarly, CLIC screens lake ice-on and ice-off dates using filters.The data are flagged if it misaligns with expectations, for example, ice-on in early October for northeastern or midwestern lakes in the USA.Participants are contacted to provide more information about their observations, and data are corrected if the error is confirmed.CLIC also has a private website where participants can access and view the data for their assigned lake(s) in tabular and graphical format.If errors are identified, participants contact us to make corrections.While errors are expected to occur, acquiring high-quality data from citizen science is possible by implementing multiple checks in an SOP to ensure errors are detected, resolved, or removed.
While developing and implementing SOPs is essential, it is equally important to document these procedures.Particularly for citizen science projects, the challenges associated with acquiring funding (Chase and Levine 2016;Turrini et al. 2018) can result in a high turnover of staff (Stepenuck and Genskow 2018).Documenting procedures are needed to keep consistency over time, especially for long-term projects.Furthermore, in a citizen science context, if the intended use of the data is for scientific research or decision-making, documenting SOPs clarifies the level of data quality available (Stepenuck and Genskow 2018).If a rigorous SOP is in place, this can help build credibility.Similarly, providing metadata files can help encourage the reuse of citizen science data by professionals (e.g., scientists) since, with this information, it is possible to decide if the data are suitable for their use (Williamson et al. 2016).
In addition, with the wider use and access to ICTs, many citizen science projects have evolved to include ICTs, such as websites and mobile applications, to submit and share data (Newman et al. 2012;Guo et al. 2014).For example, the Lake Observer application, developed by the Global Lake Ecological Observatory Network (2017), allows participants to easily submit lake and water quality observations using their mobile devices.For those with technical skills and access to ICTs, this can create an easier data submission and sharing process, encouraging public participation (Brouwer et al. 2018).However, ICTs require regular monitoring and management as they can also be a source of frustration and discouragement when they are challenging to use or do not function as intended (Carballo-C ardenas and Tobi 2016; Benyei et al. 2021).For example, problems with ICTs were one of the main factors discouraging participation in the citizen science project CONECT-e, which documented traditional ecological knowledge in a wiki-like platform (Benyei et al. 2021).At CLIC, we have a project leader responsible for monitoring the project's websites and addressing issues that arise to avoid causing frustration and discouraging participants from submitting or viewing data online.
Guideline 8: Monitor and evaluate the project Monitoring and evaluating an established citizen science project can help ensure it runs efficiently, necessary changes are made as the project progresses and evolves, and participants are satisfied with their experiences.This is another guideline the CLIC leadership team frequently revisits (Fig. 3).Since CLIC was established for decades before it was transferred to the new leadership team, it was essential to monitor the changes implemented during the transition period and ensure the participants still enjoyed volunteering, although the founder retired.In the long term, we will continue to monitor and evaluate the project closely to ensure that the goals are being met, improvements are implemented when required, and participants continue to be satisfied with the project's progress.
When monitoring and evaluating an established project, questions to consider include: are we meeting the project's goals?Can we increase participant diversity?Are the strategies for recruitment and retaining participants effective?To track how a project is operating, a monitoring and evaluation plan can be a helpful tool, especially during the project's initial planning (West and Pateman 2016).Plans can be tailored to each project and may include data on participant demographics (e. g., age, gender, and ethnicity) to track participant diversity or formative evaluations to monitor if proposed goals are being met (e.g., Patton 2002;West and Pateman 2016).The information gained from this plan can help track the changes that are necessary to improve the project as it progresses and evolves (West and Pateman 2016).
Participant satisfaction is a key component to monitor.While there is a genuine concern for participant satisfaction, there is also a practical aspect.If participants are satisfied with their role and the overall project, it can help sustain their involvement (Asingizwe et al. 2020), which is especially significant in long-term projects.Practicing reciprocity is an integral approach to ensure participant satisfaction.Different forms of reciprocity can be implemented, including recognition (de Moor et al. 2019; Asingizwe et al. 2020), awarding prizes (Dutterer et al. 2014), offering incentives (Dutterer et al. 2014), and offering learning opportunities (Cunha et al. 2017).Conducting formative evaluations (e.g., participant surveys [Patton 2002]) or conversing with participants can help determine the level of satisfaction and forms of reciprocity that motivate participation (Asingizwe et al. 2020).It can also inform project leaders if participants receive enough benefits for the time and effort they dedicate to volunteering (Adler et al. 2020), and changes can be made as needed.
For long-term participants, motivations can change, or new ones can arise as the project progresses (Ryan et al. 2001;Carballo-C ardenas and Tobi 2016).For example, for several citizen science projects in the Dutch Caribbean focused on detecting and controlling invasive lionfish, most participants were initially motivated to join to help protect the environment (Carballo-C ardenas and Tobi 2016).However, while some participants retained their desire to help the environment, new motivations also developed, including learning and developing new skills, socializing with others involved in controlling lionfish, self-enhancement, and wanting to improve the invaded areas for recreational purposes.Therefore, given the changing nature of participants' motivations, keeping track of changes in a monitoring and evaluation plan can help ensure participants' needs are met throughout the project's lifespan.
To monitor and evaluate CLIC and practice reciprocity, we started hosting an annual online conference for participants.The conference includes presentations on some of the latest lake ice research and information about how the data collected for CLIC have been used (e.g., State of the Climate reports).This conference also allows us to show our gratitude, engage in conversations with participants prompted by prepared questions, learn about participants' motivations, and receive feedback on how to provide better support and improve the project.

Guideline 9: Engage with participants and encourage participant interaction
Engaging with participants helps build rapport and trust within citizen science projects and encourages participation (Rotman et al. 2014;Turrini et al. 2018;de Moor et al. 2019).For over three decades, engagement has significantly contributed to the success of CLIC and will continue to be a priority in the long term (Fig. 3).There are different approaches for project leaders to engage with participants.For example, in the Adopt-a-River project in Brazil, project leaders provided learning activities, published online blogs, communicated with participants via email, and regularly delivered webinars (Cunha et al. 2017).This effort resulted in a higher participation rate compared to other projects, with 80% of the data collected by almost 50% of participant teams (Cunha et al. 2017), while in most projects, 80% of data is collected by 20% of participants (Franzoni and Sauermann 2014).
Engagement can also be considered a form of recognition that is essential to sustaining participation.For example, in a project focused on malaria control in Rwanda, participants considered receiving feedback from project leaders as a form of recognition (Asingizwe et al. 2020).Recognition can also help avoid the circumstance in which project leaders receive full acknowledgment for the work accomplished in a citizen science project by focusing on the participants instead.At CLIC, we highlight our participants' stories to recognize their involvement.For example, we shared an article on our website about the Great Ice Off Fishers Cove Contest in Three Rivers, Michigan, hosted by one of our participants and published by a local magazine.Also, we requested that National Public Radio interview some of our CLIC participants when they expressed interest in interviewing us about the project.Furthermore, when CLIC data are used for any research we conduct, we, at minimum, recognize participants in the acknowledgment section.As project leaders, we are responsible for recognizing participants for their work and highlighting their stories whenever possible.
Furthermore, encouraging participant interaction can help build trust in citizen science projects (Cohen 2010;Preece 2016).Indeed, participants need to trust project leaders (Benyei et al. 2021), but they also need to trust that other participants are properly fulfilling their roles.For example, Williamson et al. (2016) found that participants' willingness to share data was associated with trusting that other participants were also contributing high-quality data.Participants were concerned that the value of their data would decrease if associated with low-quality data submitted by others.Project leaders can facilitate building trust among participants by providing in-person and online opportunities to allow them to interact with one another.These opportunities can include workshops, training sessions in small groups, and online forums where participants can gain a sense of belonging and relatedness (Palacin et al. 2020).At CLIC, hosting our annual online conference and dedicating time for small group discussions allow participants to interact with one another, helping to build trust within the project.

Conclusion
Many lessons can be learned from successful citizen science projects, helping to guide the process of establishing new and sustaining existing projects.In this synthesis, we summarized the lessons learned from inheriting CLIC and researched the elements that have helped other citizen science projects succeed.The guidelines we developed provide suggestions to consider when establishing and sustaining a citizen science project by discussing topics such as diversity and inclusivity, approaches to encouraging participation, approaches to project logistics, community engagement, reciprocity, building trust, and sustaining projects long-term.While developing these guidelines, we found that successful projects have strong leadership, are well-organized, and have open and clear communication between project leaders and participants (e.g., Cooper et al. 2007;Land-Zandstra et al. 2016;Wald et al. 2016;Alexandrino et al. 2019).These three common features are the key to successfully incorporating these guidelines.
As we continue to face the consequences of global environmental change, collaborative science becomes an increasingly crucial part of understanding and addressing environmental stressors.Of particular concern are the stressors affecting aquatic ecosystems, including biodiversity loss (Witte et al. 2000), pollution (Schwarzenbach et al. 2006;Bernhardt et al. 2017), nutrient enrichment, overfishing (Hilborn et al. 2003), and changes in land use, hydrology, and climate (Woolway et al. 2020).With the urgency to further understand these stressors, there has been a proliferation of remote sensing technology and global modeling efforts, increasing the spatial coverage of lakes.However, small inland lakes are often missing from records (e.g., Sharma et al. 2020;Woolway et al. 2020) that use these technologies and, therefore, require local efforts to monitor in situ to obtain information from smaller lakes.An effective approach to obtaining missing data involves the public in the scientific process via citizen science, especially when challenged with limited resources.Fortunately, as the value of citizen science has become increasingly recognized, many more projects have been established.With the many unknown consequences of environmental stressors, citizen science provides a unique opportunity to incorporate diverse perspectives, experiences, and knowledge systems.This information gathering will be necessary to help address the consequences of global environmental change and improve the state of aquatic and other natural ecosystems.

Fig. 1 .
Fig. 1.Examples of a postcard (A, C), letter (B), and pictures of study lakes (D, E) that participants have sent to CLIC.

Fig. 2 .
Fig. 2. Map of the study lake locations where CLIC participants collect ice phenology data.Colors represent the length of the data record for each lake.

Fig. 3 .
Fig.3.Suggested guidelines to establish and sustain long-term citizen science projects.Also included are the frequencies that guidelines are revisited at CLIC.Note that different citizen science projects will likely revisit the guidelines at different frequencies.