Agriculture Reference
In-Depth Information
temperature measurement was taken. The size of the target viewed by the IRT sen-
sor is dependent on the sensor's field of view and its distance from the target. Target
emissivity, and sensor body temperature can affect target temperature measure-
ments, but temperature drift can be addressed with calibrations or temperature com-
pensation (O'Shaughnessy and Evett 2010a).
Irrigation scheduling algorithms that make use of infrared thermometry include the
Time-Temperature-Threshold (TTT) algorithm, patented as the Biologically Identified
Optimal Temperature Interactive Console (BIOTIC) for managing irrigation by the
USDA under Patent No. 5,539637 (BIOTIC; Upchurch et al. 1996). Briefly, the TTT
technique can be described as comparing the accumulated time that the crop canopy
temperature is greater than a crop-specific temperature threshold with a specified criti-
cal time developed for a well-watered crop in the same region. The TTT technique has
been used in automatic irrigation scheduling and control of plant water use efficiency for
corn in drip-irrigated plots, and soybean and cotton in LEPA-irrigated plots (Evett et al.
1996; Lamm and Aiken 2008; Peters and Evett 2008; O'Shaughnessy and Evett 2010a).
The crop water stress index, whether calculated using an empirical (Idso et al. 1981) or
theoretical basis (Jackson et al. 1981), is another thermal based stress index that can be
used with measurements from IRTs to map crop water stress (Wang and Gartung 2010),
predict yields, or time irrigations (O'Shaughnessy and Evett 2011b). Unlike the BIOTIC
method, these indices do require ancillary meteorological data in addition to measure-
ments from crop canopy temperature. Both of these automatic irrigation-scheduling
methods produced crop yields that were similar or greater than those resulting from
scientific-based manual irrigations planned by using the neutron probe.
Spectral radiometers, either handheld instruments or aerial mounted, have been used
in lieu of weather-based data to derive basal crop coefficients (
K
cb
) (Hunsaker et al. 2003,
2005) and infer crop water use for wheat (Hunsaker et al. 2007), cotton (Hunsaker et
al. 2009), and soybean (Wang et al. 2002a) using data from ground-based radiometers
to calculate the normalized difference vegetative index (NDVI). Studies by Neale et al.
(1989) and Bausch (1994) also demonstrated that NDVI values, made from ground-based
spectral radiometers, were used successfully in scheduling irrigations for corn. It is plau-
sible that spectral radiometers can be adapted onto moving sprinkler systems to monitor
crop canopy cover and crop nutrient status. The Agricultural Irrigation Imaging System
was a linear move irrigation system modified by Haberland et al. (2010) to remotely
sense spectral reflectance from the crop canopy and construct georeferenced field maps
of vegetation, nutrient, and water status.
Multispectral data can also be used to map leaf area index and detect problems in
agricultural fields, such as the presence of weeds (Aitkenhead et al. 2003; Tellaeche
et al. 2008), disease (Wang et al. 2004), soil salinity (Wang et al. 2001, 2002a,
2002b), and lack of nutrients (Bausch and Diker 2001; Clay et al. 2006; Wu et al.
2007). Weed and disease control can be rendered efficient using site-specific man-
agement if patches of the infected crop can be identified and sprays applied only to
the impacted areas. Detection of diseased patches can also lead to improved water
use efficiency with the response of withholding irrigation if disease onset occurs
early on in the growing season and the yield potential is forecasted to be less than
profitable. Geographic and spectral data locating the onset and reporting the pro-
gression of plant disease (Steddom et al. 2003, 2005) and its effects on crop yield and
