Recommending Environmental Big Data Using Semantically Guided Machine Learning - Large Scale and Big Data: Processing and Management - page 478

they mix instantaneously at a scale of hundreds of meters and whose density is
inversely correlated with the bulk soil moisture. Data from Hydroinnova CRS-1000
cosmic ray soil moisture probe [15] deployed in the Tullochgorum site in Tasmania,
Australia was used for this study. The Tullochgorum CosmOz cosmic ray sensor was
calibrated using ground truth soil samples collected around the probe. Soil moisture
was measured using the oven-drying method and area-average bulk soil moisture
was computed. The CosmOz cosmic ray probe comprising neutron detectors plus
associated electronics is manufactured by Hydroinnova, LLC, of Albuquerque, New
Mexico, USA. An Iridium satellite modem then transmits the data at 1-hour time
intervals to the CosmOz CSIRO data server (Figure 15.6). Table 15.2 represents an
example of hourly data format transmitted from the probe through the Iridium satel-
lite communications system to a CSIRO data server. Data files were made available
as .csv format. Most important and practical data attributes were atmospheric pres-
sure, neutron pulse count, and rainfall (separately measured using a rain gauge),
which have been used for this study. Time series data from the CSIRO server were
downloaded, extracted, and pressure correction on the neutron counts were per-
formed dynamically, using CosmOz web adaptor. Finally, the two time series namely
“pressure corrected neutron count” and “daily rainfall” were made available for the
data integration in the later stage, of the architecture.

FIGURE 15.6 Time series data from Hydroinnova CRS-1000 cosmic ray soil moisture probe.

TABLE 15.2
Cosmos Data Format

Time Atmospheric Pressure Neutron Pulse Count Rainfall

7/05/2011 992.3 1217 0.00