Designing Water Quality Sensors for Community Impact
Alandra Kahl outlines how to design water quality sensors for community impact.

by Alandra Kahl

This post is part of a series

In late 2014, I received an unusual voicemail. The person on the other end was a fisherman, and wanted to know about the health of the local waterway. He had found my name via our university website and thought I might be able to help. I returned his call and we spoke for about an hour about fishing, the weather and the local waterway. He was a member of Trout Unlimited and wanted to know if there was a way to monitor his local stream. At the time, I gave him a number of suggestions- he could collect a sample and bring it to campus or a local lab, or we could meet to take a look together. After I hung up the phone, it struck me that there were very few options available for citizen scientists who might be interested in monitoring their local water quality that were able to give the kind of data fisherman might be interested in and also inexpensive. That fall, my research group collaborated with the local group to create a low-cost Arduino based water sensor we would eventually name the AWQUA (Autonomous Water Quality Utility for Analysis).

This project utilized the resources of Penn State along with the input of a local community group in order to fill a gap in knowledge of local water quality. Many small waterways are not monitored, or only occasionally assessed by USGS or the EPA if there is a concern in a larger adjacent waterway.  For example, a chemical spill in the Monongahela River might affect the Youghiogheny River, which is a tributary, but only the Monongahela River is routinely monitored. By connecting with community groups of concerned citizens and helping them to become citizen scientists, we are creating a data network that will give much more comprehensive picture of our local water quality.

There are a few initial hurdles to collaboration in this way; the first being a conversation about initial design ideas. We met several times in the early stages of the process to help put together the outline of the initial capabilities for the sensor. The community and my research group worked together by creating vision boards, then breaking out those boards into specific capabilities for the sensor. This was done using large pieces of paper headed with suggestions for development, such as size, capabilities, cost, etc.  It was important to work together to translate the words of the community members from these pages into a research project. As part of the design process, student researchers broke down the community conversations into what was feasible versus what the community wanted.

For example, if a community member said they wanted to know the health of the water, we worked to break that down into sensing dissolved oxygen and water temperature. It was important to make sure ideas were heard before discussing what ideas might be feasible for the sensors. Returning to the example of the dissolved oxygen sensor, this aspect did not ultimately move forward in the design process due to the expense of inclusion of this type of sensor. To keep the project within an affordable cost (we used ~$100 per device), the number of sensors included was reduced and the housings were 3D printed, rather than purchased. The design advantage of using 3D printing to create the housings was flexibility within the design of the housing, which was an advantage in the long run. Being open about these design trade offs and the reasons behind design changes was important to helping the community feel involved and heard during the design process.

Throughout this process, undergraduate students were key liaisons between the community members and other working on the project. By connecting them to the research process through interaction with the community group, we were able to turn this project into an educational opportunity for both sides. This aspect was important to the university as well, so it was important to get buy in from Penn State to have access to funding and laboratory space.

There are a few potential hurdles to working with a university or college on a citizen science project. Community groups must connect with a researcher that has the time to focus on their project as well as the means to support that work. At Penn State, we were able to leverage several avenues of funding from the College of Engineering as well as including the project in a larger grant proposal to the National Science Foundation. Anytime something is created at a university, there is also question of intellectual property (IP). Universities like Penn State serve the public interest by disseminating knowledge through both research and teaching. While open access publication may be appropriate for some research outputs, others need to be kept confidential temporarily so that they can be patented and developed commercially. Dedicated university offices handle the business of IP. While most citizen science projects remain open source, it is important to keep this aspect in mind as a member of a community group that may want to distribute their work to others, as there may be IP concerns.

The AWQUA project is currently in the final stages of completion for dissemination back into the community. Since we have worked closely with our local group to develop the sensors, we have a pool of individuals to assist with testing them within the environment and reporting on difficulties or improvements that may need to be made to the AWQUA to make it more robust or easier to use. It is important the community groups stay in communication with the local liaisons to ensure that they are available for feedback when the project reaches a point for testing. Both parties benefit from remaining in communication through the process as they can develop the device or protocol that best meets both their needs.

From a phone call to a finished device and manual for usage, the AWQUA project has benefited from close collaboration with our local community group. We are looking forward to distributing the new devices back into the community and moving forward to collect data that is not only relevant to fishermen, but to the local environmental health of the region as a whole.

Alandra Kahl is an Associate Teaching Professor of Environmental Engineering at Penn State  Greater Allegheny who advocates for open source solutions to  environmental issues in science and engineering.