Geoscience Reference
In-Depth Information
In the past, geo-referenced EC
a
measurements have been correlated to associated yield-moni-
toring data with mixed results (Corwin et al., 2003; Jaynes et al., 1993; Johnson et al., 2001; Kitchen
et al., 1999; Sudduth et al., 1995). These mixed results are due, in part, to a misunderstanding of the
relationship between EC
a
measurements and variations in crop yield. As pointed out by Corwin and
Lesch (2003), crop yield inconsistently correlates with EC
a
due to the influence of soil properties
(e.g., salinity, water content, texture, etc.) that are being measured by EC
a
, which may or may not
influence crop yield within a particular field, and because a temporal component of yield variabil-
ity is poorly captured by a state variable such as EC
a
. Corwin and Lesch (2005a) provide a recent
review of the application of geo-referenced EC
a
measurements in agriculture with particular atten-
tion to precision agriculture applications.
Site-specific management units (SSMUs) have been proposed as a means of dealing with the
spatial variability of edaphic properties influencing crop productivity to achieve the goals of SSM.
A SSMU is simply a mapped unit of soil that is managed the same to achieve SSM goals. In a strict
sense, the task of delineating SSMUs is extremely complicated because all edaphic, anthropogenic,
topographic, biological, and meteorological factors influencing a crop's yield must be considered.
One means of simplifying the complexity of delineating SSMUs is to define SSMUs based on a
single factor, such as edaphic properties, and determine the extent of the variability of crop yield
due to the single factor.
It is hypothesized that in instances where EC
a
correlates with crop yield, spatial EC
a
informa-
tion can be used to direct a soil sampling plan that identifies sites that adequately reflect the range
and variability of various soil properties thought to influence crop yield. The objective of this study
is to utilize an intensive geo-referenced EC
a
survey to direct soil sampling and to identify edaphic
properties influencing cotton yield, and to use this spatial information to make recommendations
for SSM of cotton by delineating SSMUs based solely on edaphic properties influencing cotton
yield. This paper draws from previous more detailed work conducted and published by Corwin and
colleagues (Corwin and Lesch, 2003, 2005b; Corwin et al., 2003).
16.2
MAteRIAlS And MethodS
16.2.1 s
t u d y
s
i t e
A 32.4-ha field located in the Broadview Water District of the San Joaquin Valley's west side in
central California was used as the study site. The soil at the site is a Panoche silty clay (thermic
Xerorthents), which is slightly alkaline with good surface and subsurface drainage. The subsoil is
thick, friable, calcareous, and easily penetrated by roots and water.
16.2.2 y
i e l d
M
o n i t o R i n g
a n d
ec
a
s
u R v e y
Spatial variation of cotton yield was measured at the study site in August 1999 using a four-row
cotton picker equipped with a yield sensor and Global Positioning System (GPS) receiver. The yield
sensors measured average seed cotton yield. All subsequent references to cotton yield are with
respect to seed cotton yield. A total of 7706 cotton yield readings were collected (Figure 16.1a).
Each yield observation represented a total area of approximately 42 m
2
. From August 1999 through
March 2000 the field was fallow. On March 2000, an intensive EC
a
survey was conducted using
mobile fixed-array electrical resistivity equipment developed by Rhoades and colleagues (Carter,
1993; Rhoades, 1992). The fixed-array electrodes were spaced to measure EC
a
to a depth of 1.5 m.
Over 4000 EC
a
measurements were collected (Figure 16.1b).
16.2.3 s
a M P l e
s
i t e
s
e l e c t i of n
, s
o i l
s
a M P l i n g
,
a n d
s
o i l
a
n a l y s e s
Data from the EC
a
survey were used to direct the selection of sixty sample sites. A spatial statistics
software package, ESAP-95 version 2.01, developed by Lesch et al. (2000) was used to determine
