Skip to Content

Sensor Network at the Andrews Forest

Printer-friendly versionPrinter-friendly version
Issue: 
Fall 2012

Adam Kennedy (AND), Don Henshaw (AND)

Researchers at the H. J. Andrews Experimental Forest LTER site maintain over 30 Campbell dataloggers for collection of climatological and hydrological measurements. Currently, about half of these dataloggers use existing radio telemetry to stream data directly to a base station at the Andrews Headquarters on an hourly basis. With the retirement of long-time systems administrator Fred Bierlmaier, and the hiring of Adam Kennedy, this past year has brought changes to the way these streaming data are handled. We have moved away from customized procedures to more fully embrace capabilities within the Campbell LoggerNet (LN) software. We have upgraded our LN server workstation, added a remote LN database to host raw data, integrated LN Real-time Monitoring and Controls to handle real-time data visualizations, and began sending auto-generated alerts to personnel on site.  This updated toolset will enable the site to 1) seamlessly integrate with the LTER database located at Oregon State University (OSU), 2) immediately begin plotting highly customizable graphs online in near real-time, and 3) auto-notify information managers and field technicians of sensors that are in need of attention.

Andrews information managers were energized by information exchanges on sensor data at both the NERC/SensorNIS workshop at Hubbard Brook in 2011 and the recent ASM 2012 SensorNIS workshop at Estes Park. Site personnel are recoding stored procedures and other programs to schedule daily transformations from the raw streaming data in the LN database into our final online table formats. This transformation currently includes QC range checking and flagging of values, but other checks are being discussed. We are also restructuring our output tables to include coded attributes to indicate changes in methodology or instrumentation in collection of each measurement, and to include a quality level flag. The structure changes will also include revisions to our attribute names for better standardization and clarity of our hydrometeorological records.

In addition to these noted improvements in the workstation environment, the first phase of the new wireless communication network that extends deep into the forest - providing over 60mbps of bandwidth to researchers in remote locations – is complete. This network consists of an array of point-to-point and point-to-multipoint wireless (5.8 GHz, 2.4 GHz, and 900 MHz) links.  This wireless backbone and associated cloud radios will support long-term research, high-resolution sampling campaigns, dynamic interpretative trails, and virtual classrooms.