Dr. John Selker, hydrologist at Oregon State University and one of the scientists behind the Trans African Hydro and Meteorological Observatory (TAHMO) project, gives his perspective on the future of sensor technology.
What sparked your interest in science?
I was kind of an accidental scientist in a sense. I went into water resources having experienced the 1985 drought in Kenya. I saw that water was transformative in the lives of people there. I thought there were lots of things we could do to make a difference, so I wanted to become a water resource engineer. It was during my graduate degree process that I got excited about science.
What was the first sensor you developed?
I’ve been developing sensors for a long time. I worked at some national labs on teams developing sensors for physics experiments. The first one I developed myself was as an undergraduate student in physics. I was the lab instructor for the class, and I wanted to do something on my own while the students were busy. I made a non-contact bicycle speedometer which was much like an anemometer. I took an ultrasonic emitter, trained it on the tire, and I could get the beat frequency between emitted sound and the backscatter to get the bicycle speed.
What’s the future of sensor technology?
Right now one of the very exciting advances in technology is communication. Having sensors that can communicate back to the scientists immediately makes a huge difference in terms of knowing how things are going, making decisions on the fly, and getting good quality data. Oftentimes in the past, a sensor would fail and you wouldn’t know about it for months. Cell phone technology and the ability to run a station on a few AA batteries for years has been the most transformative aspect of technological development. The sensors themselves also continue to improve: getting smaller and using less energy, and that’s excellent progress as well.
I think the next big thing in sensing technology is how to use what we might call “semi-redundant” sensing. What often happens is that you install a solar sensor, and then a leaf or a dust grain falls on it, and you lose your accuracy. However, if you had a solar panel and a solar sensor, you could then do comparisons. Or if you were using a wind sensor and an accelerometer you could also compare data. We now have the computing capability to look at these things synergistically.
What I would say in science is that if we can get a few more zeros: a hundred times more accurate, or ten times more frequent measurements, then it would change our total vision of the world. So, what I think we’re going to have in the next few years, is another zero in accuracy. I think we’re going to go from being plus or minus five percent to plus or minus 0.5 percent, and we are going to do that through much more sophisticated intercomparisons of sensors. As sensors get cheaper, we can afford to have more and more related sensors to make those comparisons. I think we’re going to see this whole field of data assimilation become a critical part of the proliferation of sensors.
What are your thoughts on the future of sensor technology?
Get more information on applied environmental research in our