|
Cotton, Water, Nitrogen and IPM in Precision Agriculture
|
Principal
Investigator:
Robert
J. Lascano
Texas
Agricultural Experiment Station, Lubbock
Cooperators:
Lubbock:
Hong Li, Jill Booker, Kevin Bronson,
Wayne Keeling
Beaumont:
Ted Wilson
|
|
The
main purpose of this research is to establish baseline information
on the suitability and applicability of precision agriculture
practices for cotton production.
Our
initial emphasis has been on quantifying parameters that contribute
to lint yield variability across large fields.
We selected to do most of our field studies in Lamesa using
the Ag-Cares facility as a way of promoting our results and
having immediate applicability to cotton producers in this
area.
In
our initial soil sampling, on a ¼ acre grid and soil depth,
across the field it was found that nitrate-N has accumulated
and in some places exceeds 250 lbs N/A to a depth of 4 ft.
This
accumulation is the result of applying 100 lbs/A of N fertilizer
every year for the last 5 years.
Further, contributing to this accumulation has been the lack
of significant rains resulting in a N buildup in the soil
profile.
The soil survey clearly established
that it is not sufficient to just sample the surface 6" and
that deeper depths must also be considered and tested for
N content.
Classical
statistics analysis and small plot work are not appropriate
to evaluate site-specific management schemes.
Instead,
work must be done at the landscape level and using its variability
as the source of variation for the variable in question.
Further,
spatial statistical tools such as autocorrelation, crosscorrelation
and state-space analysis are appropriate to assess parameters
that contribute to crop yield variability.
|
Using
state-space analysis we demonstrated that in the AgCares farm
three variables can predict cotton lint yield variability:
These
variables are not randomly distributed and have spatial structure
and may be related up to distances of 120 ft.
Water
and nutrients accumulate in lower elevations and as a result
these sites have higher lint yields.
These equations can be used to maximize N and water application
across the field and evaluate economic returns. We
can now explain, with a large level of confidence, where in
the field and how cotton lint yield varies. These
results can also be used for the identification of management
zones within a field.
Remote sensing provides tools to
rapidly evaluate biomass and lint yield in large areas.
All cotton spectral parameters were affected by irrigation
and interaction between irrigation and N input. Spectral reflectance,
NDVI, plant water and N use, and lint yield were associated
with elevation and soil texture.
Results
from our insect work set the basis to establish site-specific
pest thresholds.
The Scout Master for cotton data management program has
proven to be a valuable tool for cotton Extension Agents.
Scout Master for cotton has the potential to improve the cost-effectiveness
of cotton production in Texas.
Quantifying
the effects of irrigation and N fertilizer levels on insect
pest densities and insect damage to cotton, and specifically
how they affect heliothis damage to cotton is valuable baseline
data for the development of site-specific information for
insect pest of cotton.
We continue to integrate our results
using physiologically based cotton models. An anticipated
component of our site-specific cotton research will be directed
towards developing decision support systems for cotton management
using variable rates and based on economic returns.
|
