There is a science that defines sustainability. At times there may also be a legal framework to define the effort. To make sustainability happen, both the scientific foundation as well as the legal framework must be accompanied with a community effort. Community can be defined in many different ways, but what all definitions have in common is the need to secure diverse modes of participation, to provide a flexible organizational structure, and to provide a collaborative data environment that bring all elements together. Without a community behind it, any goal is difficult to achieve, despite legal or scientific support. In this short article I will summarize some of our experiences gained from developing a collaborative data environment called myObservatory (or in short, mO; see myobservatory.org), and from dialogues with our partners and users.
Let’s start with community. Communities could vary in membership. A community may include a few cattle farmers in Argentina, or it could scale up all the way to national organizations of citizen scientists, like the ASPEA organization in Portugal (as discussed here), which monitors the health of the national river network with tens of thousands of kilometers and thousands of volunteers. Community can even be a global organization that supports sustainable agriculture like Savory Global. Communities both large and small need advanced collaboration tools. The level of sophistication should not depend on the scale of the institute. This is what guided our thinking in designing our collaborative data environment.
How do we make communities, small or large, work together? There may be different factors to consider. Obviously, a compelling vision is needed in order to attract participation and to maintain enthusiasm. To translate this into action, a collaborative data environment is needed, one that would allow the community to translate motivation into tangible products, and to do so, (1) it must provide the organization with flexibility in defining and accommodating user roles; (2) it must accommodate multiple and diverse modes of data entry, and (3) it should be able to maintain the credibility and integrity of the data collected, and (4) it should allow users to generate meaningful and exciting content. It must also be easy and pleasant to use, otherwise the tools themselves serve as a barrier to achieving the goal. Let’s take a look at some of these elements.
User Roles. A collaborative data environment needs to accommodate a wide range of roles, such as administrators, technicians, consultants, citizen scientists, analysts, observers, and possibly others. A user role is not just a title. A user is defined by data access privileges, and by the options to create content and provide guidelines.
Occasional users, outside of the core group of committed users, should also be accommodated. We would like to have a core group of professionals and volunteers, providing support on a regular basis. But we should also appreciate the occasional visitors who may want to inspect our work or make occasional contributions. This is especially true when reports on hazards or special events are welcome and encouraged. People with smartphones could provide a huge source of actionable information, especially when immediate action is warranted. Imagine a user taking a picture of some environmental hazard using a smartphone, which, geotagged and dated, is then immediately wired to immediately become a part of the database and a GIS display, and generating some sort of response. This accessibility empowers communities to take charge over their environment, and could keep the entire group fired up and motivated.
Collaborative efforts in support of environmental sustainability operate in a multi-stakeholder environment. This requires careful planning. I mentioned already the strict user access controls intended, among other things, to protect privacy of data. Operating of the data environment by a private entity which is not subject to Freedom of Information Act (FOIA) could be very important, as many stakeholders may be willing to share data under limiting conditions that cannot be met by government agencies, thus preventing the fear of bad publicity from deterring remediation efforts.
We also realized that not all of those who wish to use data as participants are equally motivated. What we found, however, is that in some cases users could be enticed to participate and become full-fledged contributors by being able to benefit from the collaborative data environment in ways that may be indirectly related to sustainability. Drawing from our experience in a project focused on sustainable management of a groundwater basin in California, for example: here the list of stakeholders included state and local water agencies, as well as private well-owners and volunteers. Private well owners proved to be a challenge, as they are not required, by California law, to provide data on their wells. However, we managed to make some progress here by providing benefits that are directly related to groundwater such as updates and alerts related to regional and local trends in groundwater levels. Additionally, we provided access to agriculture related functions that are indirectly related to groundwater (e.g., degree-days needed for pesticide management, tools for analysis of pumping tests). For this, we implemented a wide range of analytical and scientific tools, all mounted on a GIS platform. For example, consider this image which provides a snapshot of groundwater levels and flow directions at the Sonoma Creek Groundwater aquifer in California. This map is generated in real-time from data provided by all the stakeholders. The green dots represent groundwater state-owned monitoring wells. The blue lines represent groundwater levels and the red arrows mark flow directions. Looking at consecutive snapshots like this, one could draw conclusions about trends. Well owners are particularly worried about the water level falling below critical elevations required by their pumps. Addressing this and similar concerns could be very useful in attracting participation.
Multiple Data Entry Modes. Flexibility in defining metrics for sustainability requires having multiple modes of data entry. This could include manual data entry into specially designed forms, file uploads and sensor feeds. All data should be geo-tagged and dated. Editors should be allowed to fill in the blanks with any data that is not geo-tagged.
Smartphones are particularly useful for connecting with a large number of users in real-time. In mO, we view the smartphone as a vital component, and to accommodate it, we created seamless connectivity between smartphones and desktops. Our smartphone technology allows quick and seamless assimilation of data. Data transmitted via smartphones includes pictures, notes, and filled-in forms, and it becomes actionable as soon as it is being transmitted and displayed, which could take a fraction of a second. For example, in our Natuf Project in the Middle East, users record information on new environmental hazards by filling in specially-designated forms and taking notes and pictures. This information is transmitted and as soon as received it is used to update a hazard map in real-time. This map is then processed together with a vulnerability map using a built-in algorithm, producing an updated risk map. This process is demonstrated in the Figure below. The vulnerability map represents local conditions (soil, vegetation, water resources, land use, depth to groundwater, and others). The map at the center represents recorded hazards. The multicolored icons mark where risk were reported. These icons are clickable, to reveal all the relevant information. Each icon point contains a description of the pollution hazard and perhaps even a picture, which can be viewed on top of the risk map to understand what is causing risk in an area.
The left and center figures above represent intermediate output maps, generated by interpolation from point sources of hazard and risk data. These two maps are then processed to produce the risk map, shown on the right. The risk map updates daily. It is a vital tool for maintaining the sustainability of the underlying aquifer. This entire process is executed using scientific modules embedded in mO. The complete user-smartphone-desktop process is shown and discussed here.
Maintaining Data Integrity. I am not talking here about data quality. This is a separate issue, which can be addressed in different ways, e.g., by training and verification. What integrity means is adhering to well-known practices of data custody. That requires maintaining strict user access controls and maintenance of a chain of custody. Site administrators should be given the necessary tools to maintain the integrity of the data, and required to follow strict data protocols governing data and what can be done with it.
Content and Analytics: Making Sense of Data. This is the most rewarding component of the myObservatory collaborative data platform. Ideally one should just be able to analyze the data as it flows in, identify trends, generate or receive alerts, communicate with stakeholders, compare results across regions, identify data needs, and manage volunteers, all by clicking on an icon (preferably on a map). To make this happen, we implemented analytical tools that are used universally. We also realize that different organizations have different needs, and so we have the option to implement project-specific or even proprietary tools.
Summary. What I presented here is a short summary of an environmental information management system, tailored for the need of sustainability. It was developed by a group of engineers, programmers and sustainability experts. It evolved over many years of experience and user interaction. Give it a try and let me know what you think so that we could improve our technology. Get in touch with myObservatory if you need any assistance getting started, or want to discuss using it as a potential solution to a problem. Or, just post a comment or a question here!