|
Reporting
Period: 1 September 1997 - 31 August 2000
Objectives:
1. To
determine the water and nitrogen balance within large agricultural
fields. This involves both experimental and a modeling of the nitrogen
cycle as it relates to water and nitrogen management in relation to
cotton lint yield.
2. To determine the spatial and temporal variability of N within a
large field.
3. To assess the use of remote sensing for site-specific management
of cotton, and detection of multiple stresses in cotton.
4. Integrate the effect of N, H20, and insect injury on growth, development,
maturation, and yield of cotton.
5. Develop decision support software that predicts the site-specific
impact of multiple stresses on cotton crop growth, development, maturation,
and yield.
A. Summary
of Progress
Objective
1 and 2: Quantify the spatial and temporal variability of factors that
can be addressed by precision agriculture practices.
Topographic
land features may cause spatial variability in soil water, particle-size,
and N and P distribution resulting in variable crop yields across the
field. Topographic features such as surface elevation influence the
spatial distribution pattern of organic C, particle-size, soil water,
inorganic N, and crop yield. Surface elevation and curvature, soil organic
matter, P, and K contents strongly contributed to spatial and temporal
variability of maize yield on a hill slope in NY State, and it has been
suggested to use a topographic parameter such as surface elevation and
curvature to determine spatial patterns of yield and soil erosion. Further,
soil physical and chemical properties should be measured as a function
of space and time, and spatial-domain methods based on state-space models
are appropriate to describe spatial relationship between soil water,
nutrient and crop variables. Understanding the relationship between
soil properties, topography, and irrigation and fertilization practices
would be a basis for increasing cotton lint yield on the Texas High
Plains. There is a need of providing a complete deterministic explanation
of the interdependence between irrigation and fertilization practices
and local field conditions. It was hypothesized that plant water and
N use and cotton lint yield were related to irrigation and N supply
and topography. Objectives of the study were to (i) determine the spatial
pattern of soil water and N distribution, N uptake by cotton, and lint
yield, (ii) quantify the effects of irrigation and N fertilization on
response variables, (iii) quantify lint yield, soil property, and topography
underlying processes using state-space approach, and (iv) compare the
state-space forecasting lint yield data with the following year measured
lint yield.
MATERIALS
AND METHODS
Site characteristics
and experimental setup
The study started in May 1998 at the Lamesa Agricultural Research Farm
of Texas A&M University. Altitude at the site declines from W to
E and downward from S toward the center, then gently rolling upward
to the N. The soil is structureless, well drained, highly erodible,
and classified as an Amarillo sandy loam. In general, the soil has a
low electrical conductivity, a moderate permeability, a moderate water-holding
capacity, and a highly calcareous substratum. The experimental area
was 32 m wide and 700 m long. Treatments consisted of irrigation at
rates of 50% and 75% calculated cotton evapotranspiration (ET), and
N fertilization at rates of 0, 90 and 135 kg ha-1, respectively. There
were 5 replications of N treatments arranged in an incomplete block
design. Neutron access tubes were installed 25 m apart for each irrigation
level. Plot size was 16 1-m rows wide and 50 m long. Cotton (Gossypium
hirsutum L., 'Roundup® Ready 2326') was seeded at a rate of 16.8
kg ha-1 on 8 May in 1998 and on 10 May 1999. To ensure plant emergence,
the center pivot applied 13 mm of water immediately after seeding. Nitrogen
fertilizer (urea, 32-0-0) was fractionally applied by a chisel into
the soil at a rate of 45 kg ha-1 at emergence, bloom, and first-square.
Herbicides, insecticides and fungicides were applied according to regional
recommendations.
Irrigation
and soil water monitoring
Irrigation treatments were applied by LEPA. Water was applied according
to rain at different growth stages every 3 d. Total water applied was
242 and 323 mm in 1998, and 190 and 286 mm in 1999 for the 50% ET and
75% ET irrigation, respectively. Soil water was monitored throughout
the growing season by neutron attenuation.
Plant and
soil sampling and analysis
Latitude, longitude, and altitude at each measurement site were taken
using a DGPS receiver. Cotton rooting depth was measured in August at
three locations per irrigation level. Soils were sampled at emergence,
bloom, and harvest to determine soil NO3-N and P variability. Samples
were taken within a 0.5-m circle from each neutron access tube to a
depth of 1.8 m. Total N uptake by cotton was measured on a 10 d interval
starting two weeks after emergence until harvest. Ten plants per plot
were taken at vegetative stage and 4 plants per plot were taken during
bloom and boll stages. At each plot plant density was measured three
times at vegetative stage.
Soil samples were air-dried and sieved to 2 mm. Soil texture was measured
using the hydrometer method. The 0.1M KCl extractable NO3-N and acidified
ammonium acetate-EDTA extractable P was measured using a Technicon Auto-Analyzer.
Root length, fresh plant biomass, and dry matter of roots, leaves, stems,
bolls, and seeds were measured with samples dried at 70°C for 48
h, and ground to a 40-mesh size. Plant N concentrations were determined
using a N analyzer. Cotton plants were desiccated with Roundup®
10 d before harvest. Cotton lint was hand harvested on 5 October in
1998 and on 12 October in 1999. Each plot lint yield was determined
from four 1 x 4 m areas 3 m apart.
Mixed model,
crosscorrelation and state-space analysis
Mixed model analysis was used to determine effects of irrigation and
N fertilization on response variables. Regression relationships and
linear correlation among the soil and cotton crop variables was determined.
The spatial correlation between two variables was determined with the
crosscorrelation covariance and the crosscovariance function. Cotton
lint yield, soil water, texture, and site elevation were identified
in state-space analyses as quadrivariate first-order autoregressive
processes. We used a state-space procedure to determine the mean matrix
parameters.
RESULTS
Spatial
and temporal pattern of soil water distribution
Soil water distribution at different soil layers varied with irrigation
level, soil depth, site elevation, and slope length across the field,
as shown in Fig. 1. Water content changed consistently across the field
and water distribution pattern at all soil layers showed a dependence
on site elevation and slope length (Fig. 1). Along the 50% ET (Fig.
1ac), the soil water content in the rooting zone was higher (0.148 m3
m-3 on average) on the south-center lower position area. Water contents
changed to lower (0.123 m3 m-3 on average) on the northern extended
upslope (Fig. 1ac). Along the 75% ET plots (Fig. 1bd), movement of water
showed a similar trend and changes were consistent across the lower
area and upslope. Surface soil water content was especially low between
500-550 m on the 75% ET plots (Fig. 1bd), where the footslope area had
a higher sand content compared to the sand content on the shoulder area.
Fig.
1. Spatial pattern of soil water content at different depths in the
rooting zone (0-0.9 m) related to site elevation. Data were measured
on 24 June 1998.
In 1998,
the temporal soil water distribution pattern decreased from June to
August then increased again towards harvest in October. Analysis of
variance indicated that there was a significant difference in the temporal
assessments of the soil water contents between measurement time (date),
irrigation level, and soil depth (results not shown).
Spatial
pattern of plant water use, N uptake, and lint yield
Similar
to the soil water distribution, soil NO3-N in the rooting zone (0-0.9
m) was high on lower positions in middle of the field (Fig. 2ab). Unlike
the soil water distribution, clay content was high on the northern upslope
and shoulder area (Fig. 2ab). Total N uptake by cotton was 178 and 220
kg ha-1 in 1998, and 182 and 209 kg ha-1 in 1999 at the 50% and 75%
ET, respectively, which showed a close dependence on irrigation (Fig.
2cd). Cotton lint yield varied with irrigation level and changed within
a short distance along transects (Fig. 2ef). Mean cotton lint yields
were 704 and 962 kg ha-1 in 1998, and 801) and 924 kg ha-1 in 1999,
respectively at 50% and 75% ET irrigation level. Correspondingly, high
yields were linked to the low positions in center field areas (Fig.
2ef), where soils contained higher water content within the rooting
zone (Fig. 1). Lint yields in 1999 declined linearly from the middle
of the field to northern upslope areas, especially on the 50% ET where
the slope length was extended (Fig. 2e).
Fig.
2. Spatial pattern of soil clay (0-0.3 m) and NO3-N (0-0.9 m) measured
in July 1998, and cotton lint yield and N uptake measures at harvest
1998 and 1999.
Cross-correlation
between Lint yield, soil water, texture, and topography
There was a cyclic, positive or negative, feedback relationship between
lint yield, soil water, and elevation measured in 1998. The crosscorrelation
functions of the cotton lint yield, volumetric soil water, clay, sand,
and site elevation ranged between -0.6 and 0.6. As illustrated in the
scatter diagram (Fig. 3), the lint yield was positively cross-correlated
with soil water, but negatively cross-correlated with elevation (Fig.
3c) on the 50% ET transect at a lag distance of ± 30 m. Lint
yield was negatively cross-correlated with clay (Fig. 3b) but positively
with sand content (Fig. 3f) across a distance of 80 m. The crosscorrelation
distance between clay and elevation was ± 40 m on the 50% ET
transect (Fig. 3d), where clay content increased with elevation and
slope length.
Fig.
3. Cross-correlation function of cotton lint yield, volumetric soil
water, clay, sand, and elevation across a lag distance of ± 180
m on the 50% ET transect and 75% ET transect. Data were measured in
1998.
State-Space
cotton lint yield models
The state-space analysis quantified how cotton lint yield measured in
1998 was related to soil water, clay, and elevation in space, and how
strongly lint yield at position i was spatially based on previous measurement
of lint yield, soil water, clay or sand, and elevation at position i-1.
The multivariate state-space equations are shown in Fig. 4. The 95%
confidence limits of estimates for yields at position i given by the
state-space models are compared with measured yield values at the previous
position i-1 along the two transects in 1998. Both multivariate state-space
equations demonstrated that cotton lint yield yi, was positively weighted
on soil water, and negatively weighted on elevation at previous position
i-1 on both transects. However, lint yield yi was negatively weighted
on the clay content Ci-1 along 50% ET transect, but positively weighted
on the sand content Si-1 along 75% ET transect. Differently, 50% ET
yield at position i is much more influenced by the previous site elevation
compared to the 75% ET yield at position i.
As illustrated in Fig. 4, all lint yield measured in 1998 on both transects
were within the 95% confidence limits (Fig. 4ab). The solid lines in
the center of the shaded 95% confidence limit represent the prediction
of the state-space equations. The autoregressively predicted values
based on lint yield measured in 1998 varied with distance across the
field (Fig. 4ab), depending on soil water, texture, and elevation. Higher
lint yield was predicted on lower positions (center field area), and
lower yield could be expected on the northern side of the field, where
upslope and summit areas have less sand content.
The forecasted yield values of Yi obtained by the autoregressive state-space
model with the lint yield data measured in 1998 were then related to
cotton lint yield measured in 1999. The correlation between the lint
yield measured in 1999 and predicted values of Yi is 0.72 and 0.74 for
the 50% ET and 75% ET transect, respectively. The autoregressive state-space
model appeared to be a successful forecast tool to predict future yields
related to soil properties and field conditions.
Fig.
4. State-space cotton lint yield model at 50% ET irrigation (a) and
75% ET irrigation (b) with i, Location; Y, yield (kg ha-1); W, soil
water (mm m-1); C, clay (g kg-1); S, sand (g kg-1); E, elevation (m);
and e, model noises.
DISCUSSION
Environmental
impact on soil water and lint yield variability
The most significant characteristics in variability of cotton response
to irrigation and N fertilization were that soil water, NO3-N and particle
size distribution, N uptake and lint yield were associated with site
elevation and slope length. Irrigation amount appeared to be the most
limiting factor causing differences in lint yield and N uptake. Water
was not evenly redistributed, although irrigation was uniformly applied
across the field. Water redistribution was primarily limited by particle-size,
elevation, and slope length, which contributed to modify the spatial
pattern of surface soil water evaporation and downward movement of soil
water. As a result, the amount of soil water supply differed with soil
depths and distance. Higher cotton lint yield and N uptake were therefore
measured on lower positions, where soils contained more water received
by runoff water from higher elevations. Inversely, lower cotton lint
yield and N uptake were measured on northern upslope and summit area,
where soils were drier because of possible wind erosion and water runoff.
Weather variability also contributed to influence cotton response to
irrigation and N fertilization within years. Effect of N fertilization
was significant on lint yield in 1999. As compared to 1998, there was
an increase of 14% of lint yield with 50% ET but a decrease of 4% with
75% ET irrigation in 1999. Our results showed that changes in soil water
were more consistent than NO3-N, and were also widely controlled by
microtopography.
State-space
description of cotton lint yield variability
As a result of spatial association, the cross-correlation distance between
cotton lint yield, soil water, clay or sand content, and site elevation
was as short as 10 m, which indicated that the interdependence of soil
and cotton crop variables and topography was within 10 m. Cotton lint
yield variability, described by multivariate autoregressive state-space
models, quantified that lint yield at i position was heavily weighted
on soil water and site elevation at previous position i-1 On the 50%
ET transect, measured lint yield (1998) decreased toward the upslope
area, so that lint yield at position i was negatively weighted on yield
at the previous position i-1, and the heavily negative weight of site
elevation was therefore significant. On the 75% ET transect, lint yield
at i position was positively weighted on the yield at the previous position
i-1, since the measured lint yield tended to increase on lower and shoulder
areas. As a result, the influence of site elevation at position i-1
on the lint yield at position i was more significant on the 50% ET than
on the 75% ET, where sites are on average 0.4 m lower than sites on
the 50% ET transect.
The positive weight on soil water and negative weight on elevation indicated
that lint yield increases at position i were weighted on an increase
of soil water and decrease of elevation. Clay on the 50% ET and sand
on the 75% ET at the previous position i-1 had a significance weight
on lint yield at position i. Clay content was high on the 50% ET northern
upslope area, and sand content was important on the 75% ET center field,
where subsurface soil contained more water through water infiltration
and resulted in higher lint yield because of better root development
and water and N use.
Future cotton yield in such a large field are weighted on soil water
and topography variability as shown by the autoregressive state-space
process. Higher lint yields would be expected on lower positions where
wind-erosion hazard should be slight and accumulation of water and nutrient
is expected. Low lint yields are predicted on upslope areas, where wind
erosion hazard should be great and the soil is susceptible to loss of
water and nutrients through wind removal. In the state-space cotton
lint yield models, yields are primarily water-limited, and site elevation,
slope length, and particle-size influence water use by affecting its
distribution including transportation, infiltration, and surface water
evaporation.
CONCLUSIONS
The role
of soil water, texture, topography, and weather variability was attributed
to their influence on cotton response to irrigation and N fertilization
practices. Topography and soil texture were the key factors causing
soil water redistribution variability and created differences in lint
yield and N uptake in a large field. Irrigation had significant effects
on soil water content, lint yield, and N uptake. The consistent pattern
of soil and crop variables was measured with higher cotton lint yield
and N uptake on lower positions. Lower soil productivity was linked
to the upslope positions, where soils are more susceptible to water
and wind erosion. The crosscorrelation distance between lint yield,
soil water, clay, sand and elevation varied between 60 and 80 m. The
state-space approach identified processes underlying crop yield variability
in heterogeneous soils. Lint yield can be forecasted using a state-space
model. State-space description of crop yield variability helped to understand
the complex relationship between crop yield, soil properties, topography,
and irrigation and fertilization practices. A variable irrigation and
N fertilization adopting to soil property and topographical conditions
could contribute to improve water and N use by crop in large fields.
Objective
3: To assess the use of remote sensing for site-specific management
of cotton and water in a large field.
Water and
N availability are commonly recognized as limiting factors for cotton
production in the semiarid southern High Plains of Texas, where wind
erosion is a hazard affecting soil water and nutrient use. There is
a need for variable application of water and N related to field topographic
features such as site elevation and slope length on the High and Rolling
Plain soils. Characterization of the spatial plant/soil reflectance
pattern could be used for a variable irrigation and N rate application
across undulating fields. Soil type, plant cover, presence of moisture
and organic matter, and landscape surface roughness can impact soil
and plant reflectance in the visible red and near infrared wavelength
regions. Plant reflectance, near infrared and red reflectance ratio,
and NVDI varied with stress events, sun angles, cultivars, and irrigation
and fertilization treatments. Landscape factors caused variation in
plant and soil reflectance signal Cotton canopy ground cover and remotely
sensed scene reflectance measured in Texas and California was linearly
correlated. Plant and soil reflectance characteristics could be used
to irrigation and fertilization management in cotton. Understanding
the relationship between the spectral characteristics and the various
cropping practices used in cotton production is key for further use
of remote sensing as a tool for irrigation and fertilization management.
For this study we hypothesized that irrigation and N fertilization rates,
and topography and soil texture variability should influence water distribution
and thus cotton spectral and agronomic response. The objectives of the
study were (i) to measure cotton/soil reflectance related to different
rates of irrigation and N fertilizer, (ii) to determine cotton/soil
spectral and agronomic characteristics, and (iii) to assess influence
of soil water distribution, topography, and soil texture in cotton reflectance,
spectral vegetation index, plant water use, N uptake, and lint yield.
MATERIALS
AND METHODS
Experimental
setup and irrigation
The field study began in May 1998 at the Lamesa Agricultural Research
Farm of Texas A&M University. The experimental area was 32 m wide
and 700 m long. Altitude at the experimental site declines from S toward
the center, then gently rolling upward to the N. The soil is structureless,
well drained, highly erodible, and classified as an Amarillo sandy loam.
The experimental treatments consisted of irrigation at rates of 50%
and 75% calculated cotton evapotranspiration (ET), and N fertilization
at rates of 0, 90 and 135 kg ha-1, respectively. There were 5 replications
of N arranged in an incomplete block design. Plot size was 16 1-m rows
wide and 50 m long. Cotton (cv. 'Roundup® Ready 2326') was seeded
at a rate of 16.8 kg ha-1 on 8 May 1998 and on 10 May 1999. Fertilizer
N (urea, 32-0-0) was fractionally chiseled into the soil at a rate of
45 kg ha-1 at emergence, bloom, and first square.
Weather conditions were either extremely dry in 1998 and periodically
wet and dry in 1999. Water was applied with a center pivot using a LEPA.
Total irrigation applied was 242 and 323 mm in 1998, and 190 and 286
mm in 1999 for the 50% ET and 75% ET, respectively.
Multi-spectral
reflectance measurement
Multi-spectral plant/soil reflectance was measured using a portable
16-channel radiometer. Distance from sensors to crop surface was 2 m.
With a 28° fov the sensor viewed an area 1.0 m in diameter. Spectral
measurements were taken twice per week within a 15-30° solar zenith
angle. Multispectral reflectance was measured over harvest rows from
four areas, two were 25 m apart on row 10, and the other two were also
25 m apart on row 11 of each plot. A 0.5 x 0.5 m. Leaf area index (LAI)
was simultaneously measured using a Plant Canopy Analyzer.
Soil water
monitoring, and plant and soil sampling and analysis
To monitor soil water content, neutron access tubes were installed 25
m apart along a transect for each irrigation level. There were four
neutron access tubes per plot. Monthly soil water content was measured
by neutron attenuation. Nitrogen uptake by cotton was measured at a
10 d interval from early vegetative stage to harvest. Ten plants per
plot were taken at the vegetative stage and four plants per plot were
taken during bloom and boll stages. Plant density was measured at the
vegetative and bloom growth stage at each plot. Cotton lint was hand
harvested in each plot, and lint yield was determined from four 4-m2
areas at reflectance measurement locations. Soil samples were air-dried
and sieved to 2 mm. Soil texture was measured using the hydrometer method.
Root fresh plant biomass, and dry matter of roots, leaves, stems, bolls,
and seeds were measured with samples dried at 70 °C for 48 h, and
ground to a 40-mesh size. Plant N concentrations were determined using
a LECO Analyzer.
Calculations,
data statistics, and mapping
Volumetric soil water content was obtained by converting neutron probe
readings using field calibration equations. The NDVI was determined
from reflectance measured in the near infrared (NIR) and red (R) bands.
Effects of irrigation and N fertilizer rates on cotton spectral and
agronomic variables were tested using mixed model procedure. Regression
and correlation relationships between these spectral and agronomic variables
were determined using the General Linear Models. The NDVI map was generated
using NDVI data measured in a 2 x 6 m grid across the plots using MapInfo
6.0 software.
RESULTS
Spatial
pattern of soil water, and cotton plant reflectance
The spatial pattern of the soil water content varied with irrigation
level, soil depth, and distance across the field (Fig. 1). Surface soil
was drier than subsurface soil and soil water was distributed heterogeneously
across the field as illustrated by the volumetric soil water content
measured on 31 Aug. 1998 (Fig. 1ab). Along the 50% ET plots (Fig. 1a),
lower water contents were measured between 350-450 m where the upslope
area is located (Fig. 2a). Along the 75% ET plots (Fig. 1b), higher
soil water contents were measured between 200-400 m, which was the lower
elevation area (Fig. 2b).
Fig.
1. Spatial pattern of soil water content, soil and crop surface temperature
and reflectance at center 661 nm, 810 nm, and 1650 nm at 50% and 75%
ET irrigation level, respectively. Data were measured on 17 Aug. 1998.
Spatial
distribution of the reflectance in the visible red, near infrared (NIR)
and mid infrared (MIR) bands attributable to spectral soil and plant
properties varied as a function of spatial soil water distribution,
where the spectral measurements were taken at plant maturity on 17 Aug.
1998. Similar to the water distribution pattern, NIR reflectance was
significantly higher on 75% ET plots (46 %) than 50% ET plots (42 %).
Correspondingly, the red reflectance and MIR were significantly lower
on the 75% ET plot, than the 50% ET plots. From south to north across
the field, the consistent trend of soil water content and spectral characteristics
was that the NIR reflectance was higher, but soil and crop surface temperature
and visible red reflectance were lower where soil water content was
higher (Fig. 1).
Spatial
pattern of reflectance related to topography and soil texture
The multi-spectral curves show the soil and cotton reflectance as a
function of topography (Fig. 2). The slope extended longer in the northern
summit on the 50% ET area, and lower position extended in the middle
on the 75% ET area (Fig. 2ab). The spectral reflectance, measured at
plant maturity on 24 Aug. 1999 showed a close dependence on site elevation
(Fig. 2).
Fig.
2. Spatial pattern of soil and crop surface temperature, and reflectance
at center 661 nm, 810 nm, and 1650 nm related to site elevation in 50%
and 75% ET irrigation area, respectively. Data were measured on 24 Aug.
1999.
As compared
to the 50% ET (Fig. 2c), reflectance in the NIR band was higher, and
soil and crop surface temperature and red and MIR reflectance were significantly
lower on the 75% ET plots (Fig. 2d), showing also a similar trend as
in 1998 (Fig. 1cd).
Temporal
pattern of soil and cotton plant reflectance
The spectral curves show the soil, plant, and water reflectance as a
function of wavelength (l) and crop development (Fig. 3). In June, red
reflectance was higher (Fig. 3a), which corresponded to a higher percentage
of the exposed soil at the early vegetative stage. As cotton grew, red
radiance decreased and reflected NIR increased from June to August (Fig.
3a). The peak of NIR reflectance was measured in mid-Aug. at plant maturity.
As a result of a decrease in LAI near the time of boll opening, reflected
NIR decreased and the red increased in September (Fig. 3a).
The spectral curves also show the soil and plant reflectance as a function
of irrigation level. Dry soil (Fig. 3b) reflected highly in the red
and increased to the MIR band. The plants absorbed more blue and red
energy than green, and strongly reflected NIR energy compared to the
dry soil. Compared to 50% ET, there was a relatively higher absorption
of blue, red, and MIR energy, and higher reflection of green and NIR
energy on the 75% ET plots (Fig. 3b). Most of these differences in reflectance
can be attributed to differences in canopy density. Higher NIR reflectance
and lower red and MIR reflectance can be translated to more plant growth
with lower water stress. The slight differences in the two red and two
NIR can be attributed to the difference in the spectral width of these
bands.
Fig.
3. Temporal patterns of seasonal soil and cotton crop multispectrum
during the growing season (a), and comparison of dry soil and irrigated
cotton reflectance (b). Data were measured in 1998.
Temporal
pattern of cotton plant water use, N uptake and NDVI
The N uptake showed a S-shape growth pattern from vegetative stage to
mid September and then collapsed, as demonstrated by the measurements
taken in 1999 (Fig. 4ab). The N uptake collapse point indicated plant
defoliation on day of year (DOY) 260 (Fig. 4ab). The NDVI was also followed
a S-shaped pathway then collapsed (Fig. 4ab). In 1998 the maximum average
NDVI was 0.69 and 0.76 at the 50% and 75% ET, respectively, determined
on DOY 229. In 1999 the maximum average NDVI was 0.68 and 0.73 at the
50% and 75% ET, respectively, determined on DOY 236.
Fig.
4. Temporal patterns of the N uptake and NDVI measured in 1999.
The temporal
pattern of LAI was similar to that of NDVI. The LAI varied between 0.12
and 2.68 during the two seasons. Both NDVI and LAI increased proportionally
with increasing irrigation level (Fig. 5). The NDVI curve showed a normal
distribution pattern whose peak was situated towards the end August.
High NDVI value was a result of an increase of reflectance in the NIR
band.
Fig.
5. Temporal patterns of the NDVI and LAI measured in 1998.
Mixed effects
of irrigation on reflectance characteristics
Mean cotton lint yields were 704 and 962 kg ha-1 in 1998, and 819 and
924 kg ha-1 in 1999 at 50% and 75% ET irrigation level, respectively.
Weather variability might have caused lint yield difference within years
since the 1998 season was exceptionally dry but unusually wet in June
1999. The random effects of the mixed model variance components (block,
and block x treatments) were not significant.
Correlation
of spectral and agronomic parameters and field heterogeneity
Cotton
spectral characteristics were linked to site elevation and soil texture.
Soil and crop surface temperature, and visible and MIR reflectance were
positively correlated to site elevation and clay content, but negatively
to sand content. As a result, NIR reflectance, NDVI, and NIR and red
ratio (NIR/R) were negatively correlated to site elevation. The negative
or positive correlation coefficients indicated that high lint yield;
N uptake and plant water content were associated with low elevation
and high sand content. Red and MIR reflectance was significantly correlated
to plant water content NIR reflectance, NDVI, and the NIR/ RED ratio
were positively linked to lint yield, N uptake, and soil water.
Cotton lint yield was strongly related to NDVI. Lint yield increased
linearly with an increase of NDVI at 50% ET and at 75% ET. In addition,
spatial distribution of the NDVI status was heterogeneous across plots
(Fig. 7). As a result of a significant interaction between irrigation
and N fertilizer on near IR reflectance, NDVI was obviously higher on
90 kg N ha-1 plot (Fig. 6a) than on 135 kg N ha-1 plot (Fig. 6b). The
75% ET plots had higher NDVI values with higher frequency as compared
to the 50% ET plot. The heterogeneous distribution of the NDVI reflected
the impact of topography and soil texture on soil water distribution
and plant growth.
Fig.
6. Near infrared and red reflectance-based NDVI map. Spectral reflectance
data were measured on 15 Aug. 1998.
DISCUSSION
Soil water,
texture, and topography variability and cotton spectral characteristics
The most significant characteristics in spectral reflectance variability
were their association with soil water distribution, site elevation,
slope length, and soil texture. Water was not evenly distributed in
the soil. The amount of soil water appeared to be the most limiting
factor in spectral cotton characterization. Increasing water in the
soil reduced reflectance in the visible and MIR bands but increased
reflectance in the near IR band, which was related to an increase of
plant water content, fresh biomass, and N uptake. Therefore a decrease
of visible radiance and increase of near IR reflectance would mean higher
soil and plant water content. The association of spectral reflectance
with the topography was marked by an increase of reflectance in the
visible region and a decrease in the NIR reflectance shoulder at 810
nm on higher positions, where plant density was significantly reduced.
The plant density was higher on lower positions than on northern upslope
areas, and plant fresh biomass had the same trend with higher values
on lower areas and lower values on higher areas. In June 1999 there
was evidence of water erosion with rye grass residues and younger plants
pulled off from upslope to footslope area due to the rain. The decline
of plant density led to an increase of soil exposure. Therefore soil
and crop surface temperature and red reflectance became important, but
near IR reflectance decreased on upslope areas, where there was less
plant coverage. Inversely, near IR reflectance increased on lower positions
in center field areas, where plants were dense with more soil water.
Spectral
and agronomic responses to irrigation and N fertilization
Irrigation practice and interaction between irrigation and N fertilization
appeared to be the most limiting factors affecting cotton reflectance.
As compared to the 50% ET, irrigation at the 75% ET level supplied to
the soil an adequate amount of water since its near IR, NDVI, LAI, lint
yield, and N uptake were significantly higher. Since 1994, N fertilizer
at a rate of 190 kg ha-1 per year has been applied to this field. In
early spring, soil residual NO3-N in the rooting depth were 134 and
121 kg ha-1 in the 50% ET and 75% ET plots, respectively. Therefore
there were no significant spectral and agronomic responses to N fertilizer
due to the large amount of residual soil NO3-N. Also, variability in
rain might have caused the differences in interaction between irrigation
and N fertilization on cotton spectral response between years. In 1998
total seasonal
The most significant relations between cotton spectral and agronomic
characteristics were that NIR and reflectance varied primarily as a
function of plant fresh biomass, MIR reflectance changed with plant
water content, and the NDVI and NIR/RED ratio were also good indices
to predict N uptake, lint yield, and plant water content. Also, the
heterogeneous distribution of NDVI was a result of heterogeneous distribution
of soil water, site elevation and soil texture within a short distance
across the undulating field. Our results showed that a variable water
and N application adopted to local field conditions would be realizable
by using such a NDVI map calibrated for local conditions.
CONCLUSIONS
Irrigation
practice, interaction between irrigation and N fertilization, and variability
in soil water, texture, microtopography and weather significantly affected
the spectral reflectance properties of a cotton crop. The NIR reflectance
increased and red reflectance decreased significantly with increasing
irrigation, plant fresh biomass, and N uptake. The MIR reflectance decreased
with increasing plant water content. Soil and crop surface temperature
was cooler on the 75% ET plots, where higher plant reflectance and lower
soil reflectance were measured. Presence of soil moisture and sand reduced
reflectance in the visible and MIR band, and increased reflectance in
the NIR band. Soil and cotton surface temperature, visible, NIR and
MIR reflectance, and soil water distribution, plant water, N uptake
by cotton, and lint yield were linked to site elevation and slope length.
Red reflectance was higher on higher site positions where slope was
also extended. Slope length increased reflectance in the visible, and
MIR band. Spectral reflectance can be used with NDVI mapping to identify
spatial patterns in soil water and N across heterogeneous field. Characterization
of spatial distribution of spectral plant and soil properties can be
the basis for developing technologies for variable irrigation and N
rate application across heterogeneous field.
Objective
4 Integrate the effect of N, H20, and insect injury on growth, development,
maturation, and yield of cotton.
.
Lamesa:
The most common arthropod species found using visual leaf samples were
pests including cotton aphid, whitefly, and thrips along with green
lacewing larvae and spiders (both predators). The most common arthropods
from sweep net samples were pests including boll weevil adults and armyworm
larvae along with predators such as spiders, ladybugs, green lacewing
adults, big-eyed bugs, and nabids. Finally, boll weevil and bollworm
caused most of the injury to squares and bolls. We examined the data
using 3 different methods: (1) ANOVA to determine the effects of irrigation
and nitrogen fertilizer levels on yield, % plant nitrogen, and insect
distributions and damage, (2) multiple regression to determine possible
correlations between insect and plant parameters and cotton yield, and
(3) examination of insect spatial distributions to determine the size
of management units for variable pesticide application.
Results of ANOVA indicate the lint yield was significantly greater at
the high irrigation level than at the low irrigation level, but that
fertilizer level had no effect on yield (Fig. 1). (Significance in all
ANOVAs was set at P < 0.05.) The opposite was true for % nitrogen
in the cotton leaves, % nitrogen was higher at the low irrigation level
than at the high irrigation level, and % nitrogen increased significantly
with increasing nitrogen fertilizer levels (Fig. 2). Results for ANOVAs
on aphid density and bollworm square damage were similar as that of
% plant nitrogen; both were significantly lower at the high irrigation
level than at the low level. Also, in all cases the irrigation level
by nitrogen fertilizer level interaction was significant indicating
that nitrogen fertilizer had no effect at the low irrigation level but
did have an affect at the high irrigation level, with the medium and
high fertilizer levels having higher aphid and bollworm damage than
the low fertilizer level (Figs. 2, 3, and 4). For bollweevil boll damage
the irrigation level by nitrogen fertilizer interaction was also significant,
because at the low irrigation level bollworm damage was significantly
less at the low fertilizer level (than the medium or high levels), but
at the high irrigation level bollworm damage was less at the high fertilizer
level than at either the medium or low level (Fig. 5).
Figure
1. The effects of irrigation and nitrogen fertilizer levels on lint
yield (g) .
Figure
2. The effects of irrigation and nitrogen fertilizer levels on % plant
nitrogen.
Figure
3. The effects of irrigation and fertilizer levels on aphid density
(3/leaf).
Figure
4. The effects of irrigation and nitrogen fertilizer levels on the proportion
of squares damaged by bollworm.
We also
used multiple correlation analyses to determine which variables-plant
(height, % nitrogen); soil (pH, electrical conductivity) or landscape
(latitude, longitude, altitude, block, nitrogen fertilizer level, and
irrigation level) had the greatest impact on aphid density, bollworm
square damage, and bollweevil boll damage. Results indicated that aphid
density was significantly correlated with soil pH (r = -0.228), elevation
(r = -0.125), and irrigation level (r = -0.278) and that the overall
R2 value was very low (0.13) indicating that these factors explained
only about 13% of the variation in aphid density in cotton. Bollworm
square damage was significantly correlated with pH (r = -.116), electrical
conductivity (r = 0.229), latitude (r = 0.191), plant height (r= 0.316),
block (r = 0.24), and irrigation level (r = -0.194) with an overall
R2 value of 0.221 indicating that the factors examined explained about
22% of the variation in bollworm damage. Finally, bollweevil boll damage
was significantly correlated with pH (r = -.325), electrical conductivity
(r = 0.235), latitude (r = 0.116), longitude (r = -0.242) plant height
(r = 0.176), nitrogen fertilizer level (r = -0.129), block (r = 0.132),
and irrigation level (r = -0.666) with an overall R2 value of 0.73 indicating
that the factors examined explained about 73% of the variation in bollweevill
damage.
Figure
5. The effects of irrigation and nitrogen fertilizer levels on the proportion
bolls damaged by boll weevils.
Finally,
we examined the distribution patterns of aphids, bollworm damaged squares,
and bollweevil damaged bolls using different sized sampling units to
determine whether insect clumping would allow for the use of insecticides
in only portions of the field. For aphids we found that if 150 m by
24 row areas (blocks) were examined to determine if the aphid density
was above the spray threshold (> 50 / leaf), insecticides would only
need to be applied to 50% of the field. If smaller areas, 50 m by 24
rows, are examined than only 38% would be sprayed, while if even smaller
areas (25 m by 12 rows) are examined the difference becomes small (37%)
indicating that 50 m by 24 row areas are the best size for aphid management
units. This is somewhat different for bollworm damage, for 50 m by 24
row areas only 50% of the field would need be sprayed; for the 25 m
by 24 row areas 45% would be sprayed, and for the smallest area 30%
would need to be sprayed. Bollweevil damage was similar to bollworm,
using the large areas, 100% would need to be sprayed, if the medium
sized areas were used as management units then 88% would need to be
sprayed, while if the small areas were used then only 58% would need
to be sprayed. For bollworm and bollweevil, the smallest area would
be the best size for the management unit.
In conclusion, our findings indicate that variable irrigation and nitrogen
fertilizer used in precision agriculture will affect insect pest densities
and damage to cotton fruits. Also, because nitrogen fertilizer and irrigation
levels affect pest densities, both can be used in a model for predicting
pest densities on a large scale. However, on a smaller scale, pest densities
were only slightly correlated with plant, soil, and landscape factors
indicating that these factors alone will not be useful in predicting
pest densities. Finally, the clumped distribution pattern of insects
and insect damage indicates that it should be possible to apply pesticides
to only portions of a field, thus allowing the survival of natural enemies
and the slowing of insecticide resistance evolution in unsprayed areas.
Lubbock:
ANOVA results for lint, seed, and carpel weights were similar; therefore,
we'll just discuss lint and seed weights (per 2 m-row treatment plots)
rather than repeating the same information for all three. Both irrigation
level and damage type (real or simulated) had significant effects on
yield. Yield was greater at the high irrigation level (means = 56.8
mg lint /plot and 90.9 mg seed /plot) than at the low irrigation level
(means = 37.8 mg lint /plot and 58.2 mg seed /plot). Yield was also
greater for real damage (means = 48.8 mg lint /plot and 76.9 mg seed
/plot) than for simulated damage (means = 45.8 mg lint /plot and 72.3
mg seed /plot). The latter was probably due to high bollworm mortality
in real damage plots. This caused yield to be greater in real damage
plots than in simulated damage plots, because we didn't take bollworm
mortality into account when simulating bollworm damage levels at the
three densities. This was done to ensure that we obtained some information
on damage even if high mortality in the real damage plots caused treatment
densities to be indistinguishable.
There were significant interactions between irrigation level and density
and between irrigation level and infestation period (Fig. 6 and 7).
Neither density nor infestation period significantly affected yield
at the high irrigation level (Fig. 6). However, at the low irrigation
level, yield decreased with increasing bollworm density (Fig. 7). Also,
yield was significantly less for the last infestation period (boll maturation)
than at either of the earlier infestation periods (Figs. 6 and 7).
We have 4 main conclusions from the Lubbock bollworm fruit consumption-injury
study. First and most simply increasing the level of irrigation increases
yield. Second, simulated bollworm damage effects yield significantly
more than real bollworm damage when bollworm mortality is not included
in simulations. Third and most importantly, increasing bollworm density
decreases yield when irrigation is low but has no effect when irrigation
is high. Finally, bollworm damage that occurs during the boll maturation
period decreases yield (in comparison with earlier cotton developmental
periods) when irrigation is low, but not when irrigation is high.
Figure
6. The effects nitrogen fertilizer level, damage level, and damage timing,
on lint yield in cotton at the high irrigation level.
Figure
7. The effects nitrogen fertilizer level, damage level, and damage timing,
on lint yield in cotton at the low irrigation level.
Leaf nitrogen
content was estimated using a chlorophyll (SPAD) meter to measure the
concentration of chlorophyll in the leaves. Results from comparison
of Dr. Chilcutt's chlorophyll (SPAD) meter readings and Dr.'s Li and
Lascano nitrogen analysis indicate that SPAD measurements were not a
good indicator of cotton plant nitrogen content. Also, mean SPAD readings
were similar for the 3 nitrogen fertilizer levels, with no significant
differences between levels. However, SPAD readings did vary significantly
with time. After an initial decrease in SPAD readings from June to July,
SPAD readings increased steadily from July to September. Finally, SPAD
readings were not correlated with yield or with insect densities or
insect damage to cotton squares or bolls, indicating that the SPAD meter
is not a useful instrument for measuring crop variability or insect
variability in cotton.
Objective
5 Develop decision support software that predicts the site-specific
impact of multiple stresses on cotton crop growth, development, maturation,
and yield
College
Station and Beaumont:
A major aspect of our research/extension program has focus on the development
of decision support software for cotton. The resulting software, Scout
Master for Cotton, has been developed to quickly and efficiently summarizes
field scouting data for electronic delivery to producers, private consultants,
agribusiness, university scientists, and other agencies and organizations
that use IPM information. Scout Master currently allows Extension Agents
to provide information via newsletters on pest, beneficial, and crop
status, and pesticide, water, and fertilizer use across a large number
of fields or across a series of sample dates for any field in Scout
Master's database. It also allows for computer access to large quantities
of organized field data that can form a readily accessible historical
database. This allows pest population development to be compared between
and among years, for different regions, and for different pest management
methods. As Scout Master's use continues to expand, the data will become
useful for evaluating resistance management programs, evaluating regional
and area-wide management approaches, and providing valuable information
on the region-level movement of potentially damaging pest populations.
Scout Master is also a valuable tool for training new Extension Agents
on how to systematically sample fields in terms of monitoring pest species
and crop growth and development, how to store data, and how best to
deliver data to growers and crop consultants. The speed and degree with
which sound pest management practices are adopted by crop consultants
and growers in a crop production regions is greatly determined by the
quality of the regions Extension Agents.
During 1997, Scout Master for Cotton was tested on about 300 fields
by extension agents-IPM. Based on suggestions made by extension agents
and their scouting teams, numerous changes and additions were made and
version 2.0 of Scout Master for Cotton was released in spring 1998.
Additions included expansion of the graphics options and inclusion of
pesticide, water, and fertilizer record keeping. During 1998, Scout
Master for Cotton was tested on nearly 3,000 fields, in the Rio Grande
Valley, the Coastal Bend, the Southern Rolling Plains, and the High
Plains of Texas. Follow-up meetings produced requests for additional
improvements, which are being incorporated and release of version 3.0
occurred in June 1999. These latest additions included options for improved
field sorting to facilitate ease of data entry, options for sorting
data to enable comparison of insect abundance as affected by planting
date, irrigation management, and plant variety, an expanded export utility,
and a utility to import geo-referenced data from global positioning
systems (GPS). Our current efforts with Scout Master for Cotton's include
several additions, including 1) a weed management component, 2) integrating
pest thresholds, and 3) incorporating PDA (Palm pilot) technology to
greatly expanding Scout Master's ease of use, and speed with which it
can be used. We have investigated the use of voice-recognition software
but have found the error rate for existing software to be too expensive
and multiple licenses excessively expensive.
Additions and improvements have rapidly expanded Scout Master for Cotton's
data management options and its ease of use. With each new season, the
program will be tested across an increasingly greater area of Texas.
The Scout Master development team will continue to meet a minimum of
twice per year, once prior to crop planting and once following harvest,
to summarize needed corrections and desired additions. Corrections and
additions will continue to be incorporated into Scout Master for Cotton
as requested.
In addition to our regular Scout Master meetings, the Scout Master for
Cotton developmental team has met with cotton experts to develop crop
stage specific control thresholds for each of the major pest species.
The incorporation of PDA-based data entry and Scout Master database
synchronization has the potential to greatly increase the speed with
which field scouting data can be entered into Scout Master.
The net result of this component of our project will be improved decision
making by growers resulting in increased net profits and decreased risks
to human health and the environment. Results of the project will also
facilitate agribusinesses in providing improved service to agricultural
customers and serve as an evaluation tool for regional and area wide
management efforts. Further developments will result in the creation
of computerized databases that will serve as valuable resources that
can be used to optimize crop production and management inputs.
B. Education/technology
transfer
·
Held a workshop titled "State-Space Analysis and Other Statistics
and its Application to Precision Agriculture" at the Lubbock center,
30 Nov - 2 December, 1998. Dr. D. R. Nielsen and Dr. Ole Wendroth instructed
the workshop. R. J. Lascano organized the workshop.
·
Held a meeting with Dr. Basil Acock from the USD-ARS crop simulation
unit to implement the use of their cotton simulation model CPM (Cotton
Production Model). This model is geared towards the management of water
and nitrogen and fits within our objectives. We are currently evaluating
the model and expect that it will be used as a management tool.
·
The CPM model is not adequate for the environmental conditions of the
Texas High Plains and we are now using the Cotton2K model from A. Marani.
We are currently evaluating this model. This model will provide a management
tool that can be used for site-specific management.
C. Milestones
Achieved:
·
During the 1998 and 1999 growing seasons, lint yield did not respond
to nitrogen fertilizer due to the high content of residual nitrate-nitrogen
in the soil. This indicates that the current recommendation of only
sampling the surface 6 inches for soil nutrient management is erroneous
and can lead to over fertilization and unnecessary cost.
·
Results from both growing seasons indicate that it is possible to estimate
plant leaf area and plant biomass from plant reflectance measurements.
The reflectance measurements are made with a hand-held radiometer.
·
Our state-space analysis indicates that it is possible to use this statistical
approach to identify management units within a field. Results indicated
that we can explain lint yield variability with a 95% confidence using
state-space analysis. This analysis shows that in Lamesa lint yield
variability is related to irrigation amount, elevation and soil N-NO3
content. These results are significant because they indicated that lint
yield for a given location can be maximized using the state-space equations.
For example, irrigation water in combination with N fertilizer can be
variably applied on the field to maximize lint yield. Also noteworthy,
is that in Lamesa NO3 content in the soil is not related to yield due
to its very high concentration.
·
Variability in site elevation, slope length, and soil texture may affect
crop response to irrigation and fertilization. A two-year (1998-1999)
study was conducted in a center pivot irrigated cotton field on the
semiarid Southern Texas High Plaines to determine cotton lint yield
and N uptake pattern related to irrigation, N fertilization, and field
conditions. Treatments consisted of irrigation at 50% and 75% calculated
cotton evapotranspiration (ET), and N input at rates of 0, 90, and 135
kg ha-1 arranged in an incomplete block design. Soil, water and crop
variables were measured as a function of irrigation, fertilization,
and space along transects. Higher soil water content, cotton lint yield,
and N uptake were linked to lower positions in the field. Mixed model
analysis showed that main effects of irrigation were significant on
increase of soil water content, plant water content, N uptake, and cotton
lint yield (P > 0.0425-0.0012). The N input had no effect in a dry
year (1998). The model residual was significant on all measured variables
(P > 0.0001). In 1998, lint yield, soil water, clay, sand, and elevation
were crosscorrelated at a lag distance of ± 30-40 m. The multivariate
autoregressive state-space analysis quantified the underlying processes
of cotton lint yield, soil water, clay, sand, and topography. Lint yield
measured in 1999 was correlated to the forecasted data by the state-space
model in 1998. Results showed that state-space description of crop yield
variability helps to understand interdependency between irrigation,
fertilization, and field heterogeneity.
·
Variability in cotton multispectral reflectance could be attributed
to irrigation and N fertilization practices and the impact of microtopography
on water and N redistribution. A two-year (1998-1999) study was conducted
in a center pivot irrigated field on the semiarid South Texas High Plains
to determine cotton spectral reflectance characteristics related to
irrigation, N fertilization, and field conditions. Treatments consisted
of irrigation at rates of 50% and 75% calculated cotton evapotranspiration
(ET), and N fertilization at rates of 0, 90 and 135 kg ha-1 arranged
in an incomplete block design. The composite reflectance properties
were investigated over a wavelength range of 447-1752 nm. Near infrared
reflectance increased and visible red reflectance decreased with increasing
irrigation. Reflectance in the red and mid infrared bands increased
with site elevation. All cotton spectral characteristics were significantly
affected by irrigation (P > 0.0020-0.0488), and interaction between
irrigation and N input (P > 0.0001-0.0163). Spectral reflectance,
normalized difference vegetation index (NDVI), plant water and N use,
and lint yield were strongly associated with site elevation and soil
texture. It is concluded that irrigation rates and landscape attributes
caused variation in reflectance signal.
D. Publications:
·
Lascano, R. J., R. L. Baumhardt, S. K. Hicks and J. A. Landivar. 1998.
Spatial and temporal distribution of surface water content in a large
agricultural field. 4th International Conference on Precision Agriculture,
19-22 July, 1998. St. Paul, MN, USA, Part A: 19-30.
·
Li, Hong, R. J. Lascano, E. M. Barnes, and P. Waller. 1999. Multispectral
remote sensing related to water and nitrogen use in cotton. 1999 Annual
Meetings American Society of Agronomy, Salt Lake City, UT, 31 October
- 4 November, 1999, p 21.
·
Li. Hong and R. J. Lascano. 1999. State-space approach for management
of field heterogeneity in cotton. Annual Meetings American Society of
Agronomy, Salt Lake City, UT, 31 October - 4 November, 1999, p 202.
·
Li, Hong, R. J. Lascano, Jill Booker, L. Ted Wilson, and K. F. Bronson.
2000. Cotton lint yield variability in a heterogeneous soil at a landscape
level. Soil & Tillage Research 1553:1-14 (In press).
·
Li, Hong, R. J. Lascano, Jill Booker, K. F. Bronson, E. Segarra, E.
M. Barnes, and L. T. Wilson. 2000 Spectral reflectance characteristics
of cotton related to soil water and topography variability. 5th International
Conference on Precision Agriculture, 16-19 July, 2000 St. Paul, MN,
USA (In press).
·
Li, Hong, R. J. Lascano, Jill Booker, K. F. Bronson, L. T. Wilson, and
E. Segarra. 2000. Underlying field heterogeneity on water and n use
in cotton: State-space analysis. 5th International Conference on Precision
Agriculture, 16-19 July, 2000 St. Paul, MN, USA (In press).
·
Barnes, E.M., T. R. Clarke, S. E. Richards, P. D. Colazzi, J. Haberland,
M. Kostrzewski, P. Waller, C. Choi, E. Riley, T. Thompson, R. J. Lascano,
Hong Li, and M. S. Moran. 2000. Coincident detection of crop water stress,
nitrogen status and canopy density using ground-bases multispectral
data. 5th International Conference on Precision Agriculture, 16-19 July,
2000 St. Paul, MN, USA (In press).
·
Li, Hong, R. J. Lascano, E. M. Barnes, Jill Booker, L. T. Wilson, K.
F. Bronson and E. Segarra. 2000. Multispectral reflectance characteristic
of cotton related to soil water, texture, and topography. (Paper submitted
to the Agronomy Journal).
·
Lascano, R. J. and H. Li. 2001. Precision farming to improve water use
(Invited paper to be published in the Encyclopedia of Water Science,
in review).
·
Li, Hong, R.J. Lascano, J. Booker, L. T. Wilson, and K. F. Bronson.
2000. Landscape-scale assessment of soil texture, water, plant reflectance,
and lint yield variability in irrigated cotton. Annual Meetings American
Society of Agronomy, Minneapolis, MN, 5-9, November, 1999, p 324.
· Wilson, L. T., R. Huffman, T. Fuchs, J. Wang. 1997.Master cotton.
Versions 1.0, 1.2, 1.3 for Windows 95/NT systems. Copyright, Texas A&M
University.
· Huffman, R., T. Fuchs, L. T. Wilson, J. Wang, M. Wallace, B.
Baugh, R. Minzenmayer, and John Norman. 1998. "Scoutmaster":
New cotton insect data entry and analysis software for Windows 95. 1998
Proceedings Beltwide Cotton Production Research Conferences. pp. 168-170.
· Wilson, L. T., R. Huffman, Mike Wallace, and T. Fuchs. 1998
Scoutmaster for cotton. Versions 2.0 for Windows 95/NT systems. Copyright,
Texas A&M University.
· Knutson, A., and L. T. Wilson. The beat bucket: A rapid reliable
method for sampling predatory insects and spiders in cotton. 1998 Proceedings
Beltwide Cotton Production Research Conferences. pp.
E. Precision
agriculture proposals:
·
Remote Sensing and in-situ Measurement of Soil Water in Precision Agriculture:
Integration of Measurements with Simulation Models. Robert J. Lascano.
Amount requested $339,000 for three years. Proposal submitted to the
Idaho National Environmental and Engineering Laboratory (INEEL). (Proposal
was funded for three years). 1997-1999.
· Remote Sensing Detection of Multiple Biotic and Abiotic Stresses
in Precision Agriculture. R. J. Lascano, PI with the collaboration of
C. M. Rush, G. J. Michels, T. L. Archer, K. F. Bronson and L. T. Wilson.
Proposal submitted to NASA for $598,271 for three years. (Proposal was
not funded), 1998.
·
Water Management of Cotton in Precision Agriculture: Lint Yield as a
Function of Site-Specific Irrigation. Robert J. Lascano. Amount requested
$60,000 for one year. Proposal submitted to the Cotton State Support
Committee. (Proposal was funded for three years), 1999-2001
·
Data set for Cotton Production Model. Robert J. Lascano. Amount requested
$4,330 for one year. Proposal submitted to the USDA-ARS. (Proposal was
funded), 1999.
·
A state-space approach to manage low energy precision application (LEPA)
irrigation systems. Robert J. Lascano, and D. Porter. Amount requested
$182,500 for two years. Proposal submitted to the Advanced Research
Program, Texas Higher Education Coordination Board. (Proposal was not
funded). 1999.
·
Yield Tracker: A yield mapping and prediction information delivery system.
S. J. Maas, R. J. Lascano and D. Cooke. Proposal submitted to the IFAFS-
USDA-CSREES. (Proposal was funded at $800,000 for three years).
· Huffman, Wilson, 1996-1997. Development of a geo-referenced
component for Scout Master for Cotton, and development of Scout Master
for Corn. TPMA. $40,000. (proposal was funded)
· Wilson, Fuchs, 1998-1999. Development of a database component
for Scout Master for Cotton. TDA. $25,000 (Proposal was funded)
· Wilson, Fuchs, 1999-2000. Incorporating Voice-Recognition,
Insect Thresholds, and Weed Monitoring into Scout Master for Cotton.
TDA $31,000 (Proposal was funded).
· Wilson, Norman, et al. 1998-1999. Incorporating voice recognition,
insect thresholds, and weed monitoring into Scout Master for Cotton.
TPMA. $41,000. (Proposal was not funded)
· Wilson, Fuchs, 1998-2000. Scout Master for Cotton. TDA. $20,000
(Proposal was not funded).
· Wilson, Fuchs, et al. 1999-2000. A multidisciplinary approach
to integrating multiple species thresholds into a holistic cotton management
system. Southern Region IPM. (Proposal was not funded).
F. Precision
Agriculture meetings attended/papers (posters) presented:
·
Lascano, R. J., R. L. Baumhardt, S. K. Hicks and J. A. Landivar. 1998.
Spatial and temporal distribution of surface water content in a large
agricultural field. Paper presented at the 4th International Conference
on Precision Agriculture, 19-22 July, 1998. St. Paul, MN, USA, Part
A: 19-30.
· Machado, S. E. D. Bynum Jr., D. T. Rosenow, G. C. Peterson,
T. L. Archer, R. J. Lascano, K. Bronson, and E., Segarra. 1999. Spatial
variability of sorghum yield: site-specific interactions of soil water
and pests. 1999 Poster presented at the Sorghum Conferences, Tucson,
AZ
· Li, Hong, R. J. Lascano, E. M. Barnes, and P. Waller. 1999.
Multispectral remote sensing related to water and nitrogen use in cotton.
Poster presented at the 1999 Annual Meetings American Society of Agronomy,
Salt Lake City, UT, 31 October - 4 November, 1999, p 21.
· Li. Hong and R. J. Lascano. 1999. State-space approach for
management of field heterogeneity in cotton. Poster presented at the
1999 annual Meetings American Society of Agronomy, Salt Lake City, UT,
31 October - 4 November, 1999, p 202.
· Lascano, R.J., J. Booker, H. Li, K. Bronson, W. Keeling, T.
Wheeler, J. Bordovsky, T. Wilson, S. Searcy, E. Segarra, J. Gannaway,
R. Boman, P. Dotray, N. Hopper, and H. Kaufman. 2000. Cotton precision
agriculture work in the Texas High Plains. Paper presented at the 2000
Cotton Beltwide meetings in San Antonio, TX 4-8 January, 2000.
· Lascano, R.J., J. Booker, H. Li, K. Bronson, W. Keeling, T.
Wheeler, J. Bordovsky, T. Wilson, S. Searcy, E. Segarra, J. Gannaway,
R. Boman, P. Dotray, N. Hopper, and H. Kaufman. 2000. Precision agriculture
in the Texas High Plains. Paper presented at the 2000 Cotton Beltwide
meetings in San Antonio, TX 4-8 January, 2000.
· Li, Hong, R. J. Lascano, Jill Booker, K. F. Bronson, E. Segarra,
E. M. Barnes, and L. T. Wilson. 2000 Spectral reflectance characteristics
of cotton related to soil water and topography variability. Paper presented
at the 5th International Conference on Precision Agriculture, 16-19
July, 2000 St. Paul, MN, USA (In press).
· Li, Hong, R. J. Lascano, Jill Booker, K. F. Bronson, L. T.
Wilson, and E. Segarra. 2000. Underlying field heterogeneity on water
and n use in cotton: State-space analysis. Paper presented at the 5th
International Conference on Precision Agriculture, 16-19 July, 2000
St. Paul, MN, USA (In press).
· Barnes, E.M., T. R. Clarke, S. E. Richards, P. D. Colazzi,
J. Haberland, M. Kostrzewski, P. Waller, C. Choi, E. Riley, T. Thompson,
R. J. Lascano, Hong Li, and M. S. Moran. 2000. Coincident detection
of crop water stress, nitrogen status and canopy density using ground-bases
multispectral data. Paper presented at the 5th International Conference
on Precision Agriculture, 16-19 July, 2000 St. Paul, MN, USA (In press).
· Li, Hong, R.J. Lascano, J. Booker, L. T. Wilson, and K. F.
Bronson. 2000. Landscape-scale assessment of soil texture, water, plant
reflectance, and lint yield variability in irrigated cotton. Poster
presented at the annual Meetings American Society of Agronomy, Minneapolis,
MN, 5-9, November, 2000, p 324
· Li, Hong, R. J. Lascano, E. M. Barnes, and P. Waller. 1999.
Multispectral remote sensing related to water and nitrogen use in cotton.
1999 Poster presented at the Annual Meetings American Society of Agronomy,
Salt Lake City, UT, 31 October - 4 November, 1999, p 21.
· Li. Hong and R. J. Lascano. 1999. State-space approach for
management of field heterogeneity in cotton. Poster presented at the
Annual Meetings American Society of Agronomy, Salt Lake City, UT, 31
October - 4 November, 1999, p 202.
· Lascano R. J. H. Li, and Jill Booker. 2000. Assessing soil
and cotton lint yield variability on a landscape-scale. Keynote paper
presented at the First Joint Congress of the Soil Science Society of
America and the German Soil Science at Osnabrück, Germany 18-22
September, 2000.
· Lascano, R. J. Water Management of Cotton in Precision Agriculture:
Lint Yield as a Function of Site-Specific Irrigation. Paper presented
at the Cotton State Support committee, Lubbock 7-8 December, 2000.
· Wilson, Member, USDA/NRI Agricultural Systems Panel, 1997-1998.
· Wilson, Panel Manager, USDA/NRI Agricultural Systems Panel,
1998-1999.
· Wilson, Scout Master for Corn demonstration, Texas IPM Technical
Advisory Committee meeting, October 1998.
G. Other
Developments:
|