.|  Baltimore Ecosystem Study
Wireless sensor networks for soil monitoring

Figure 1. EnviroMote, the second generation monitoring node.
Having accurate environmental data with a good spatial and temporal resolution is simply not possible using the traditional manual techniques. Inexpensive wireless sensors provide a completely new way to attack these problems: our experiment measures soil moisture and temperature at several depths at a few meters spatial resolution every 10 minutes, without disturbing the environment that we want to study. Combined with other meteorological data like temperature, rainfall, humidity and topographic information the system has an extremely detailed picture of the conditions in the soil at the relevant spatial scales.

Figure 2. Deployment of EnviroMotes at Cub Hill. At each location soil temperature and soil moisture measurements are taken at 10 and 20 cm depths.
In January 2005 we started to build a wireless sensor network for soil monitoring, and in summer we deployed the first system. In April 2006 the second system was deployed in Leakin park of the BES permanent forest plots. Building the system was a very useful experience. First, it became obvious that the custom programming of the wireless "motes" was a non-trivial problem, and it required more computer science expertise than most biologists have. Also, it soon became clear that the road from raw data to a calibrated spatio-temporal data set suitable for the final scientific analysis is long and cumbersome. Third, it became clear that the traditional analysis data analysis tools and processes were not adequate to these data volumes. Fourth, it became clear that we need to redesign the hardware, which also meant new software development.

Figure 2. Deployment of EnviroMotes at Cub Hill. At each location soil temperature and soil moisture measurements are taken at 10 and 20 cm depths.

Our second generation system is based on the Telos SkyMote platform, which has lower power consumption and better sleeps mode properties (Figure 1). We have designed our own 4-channel analog interface board with its own stabilized reference voltage. We designed a new watertight enclosure with a lithium battery that has a lifetime of about 2 years. Sensors are attached through a waterproof multipole connector, making field replacements much easier. The humidity inside the box is also monitored with its own sensor. . The wireless sensor network is built on top of the open source TinyOS operating system, and uses a self-routing network for the wireless connections. The wireless "motes" connected to the sensors are normally asleep, saving power, they only wake up for turning on the sensors and do a burst of data collection, then go to sleep again. At regular intervals they wake up and upload the data to the relay nodes. These, in turn are connected to the gateways, typically a relatively low power PC, connected to the internet. The data management system is using Microsoft SQL Server as the underlying database, while the preprocessing and transformation of the data is done using a set of Linux tools. The system collects temperature, moisture and CO2 concentration data belowground.

Figure 4. Soil moisture changes over four months at Cub Hill. Data from a subset of the motes are shown. Created by J. Gubchup, JHU.
Over the past five years we have deployed this system in an urban residential area (Cub Hill testbed: 53 sampling locations in urban forest and lawn habitats, Figure 2), in a mid-Atlantic deciduous forest (Smithsonian Environmental Research Center, MD: 37 sampling locations), in an agricultural field (USDA Beltsville Agricultural Research Center Farming System Project: 22 sampling location in no-till crop fields), in a high altitude desert (Atacama Desert, Chile: 3 sampling locations), and in the Amazon basin (Yasuni National Park, Ecuador: 12 sampling locations).
The data allow us to explore spatial variability of soil conditions (Figure 3) as well as examining the soil system during transient events (Figure 4).
To disseminate the soil data collected by our wireless sensor network we developed a user-friendly web interface. "Grazor" is a website for visualizing and exploring data and is available from http://grazor.cs.jhu.edu/. Using Grazor, users can select and view time series of sensor measurements across different sensor types and create 'tags' that they can share with other users of the system. Data can be downloaded either as csv or excel files.
Szlavecz K, Terzis A, Szalay A, Mus?loiu-E R, Gupchup R, Carlson D, Pitz SL, Bernard ML, Xia L, Chang M 2013. Wireless sensor network for in situ soil moisture monitoring: lessons from several deployments. Vadose Zone Journal, under review
Savva Y., Szlavecz K., Carlson D., Gupchup J., Szalay A., Terzis A. 2013. Spatial patterns of soil moisture under forest and grass land cover in a suburban area, in Maryland, USA. Geoderma 192: 202-210 DOI:10.1016/j.geoderma.2012.08.013
Terzis, A., R. Musăloiu-E., J. Cogan, K. Szlavecz, A. Szalay, J. Gray, S. Ozer, M. Liang, J. Gupchup, R. Burns. "Wireless Sensor Networks for Soil Science". International Journal of Sensor Networks. Volume 6, Issue 4, 2009. Pages 1-17.
Savva Y., Szlavecz K., Pouyat R., Groffman P., Heisler G. 2010. Land use and vegetation cover effects on soil temperature in an urban ecosystem. Soil Science Society of America Journal 74. doi:10.2136/
Musăloiu-E, R. A. Terzis, Szlavecz, K., , A. Szalay, J. Cogan, R, J.Gray Life under your feet: Wireless sensors in soil ecology. Proceedings of the Third Workshop on Embedded Networked Sensors (EmNets 2006) May 30-31, 2006 Harvard University, Cambridge, Massachusetts, USA p. 51-55.
Szlavecz, K, A. Terzis, S. Ozer, R. Musăloiu-E., J. Cogan, S. Small, R. Burns, J. Gray, A. Szalay. Life Under Your Feet: An End-to-End Soil Ecology Sensor Network, Database, Web Server, and Analysis Service. http://research.microsoft.com/research/pubs/view.aspx?msr_tr_id=MSR-TR-2006-90
Stuart Ozer; Alex Szalay; Katalin Szlavecz; Andreas Terzis; Razvan Musăloiu-E.; Joshua Cogan 2006 Using Data-Cubes in Science: an Example from Environmental Monitoring of the Soil Ecosystem ftp://ftp.research.microsoft.com/pub/tr/TR-2006-134.pdf
For more information on the sensor project and online data please go to: http://www.lifeunderyourfeet.org/
Contact: Katalin Szlavecz: szlavecz@jhu.edu
Katalin Szlavecz, Dept. of Earth and Planetary Sciences
Andreas Terzis, JHU Dept. of Computer Science
Alex Szalay, JHU Dept. of Physics and Astronomy
Gordon Heisler, John Hom, Richard Pouyat, USDA Forest Service
Yulia Savva, Towson University
Lijun Xia, JHU Dept. of Earth and Planetary Sciences (now at University of Delaware)
Jayant Gupchup, JHU Dept. of Computer Science (now at Microsoft Research)
Scott Pitz, JHU Dept. of Earth and Planetary Sciences
Doug Carlson, JHU Dept. of Computer Science