![[ Visit Client Website ]](images/banner.jpg)
Thursday, January 9, 2020: 2:00 PM
JW Grand Salon 5 (JW Marriott Austin Hotel)
Imagery from airborne systems such as Unmanned Aircraft Systems (UAS) or Satellites when
properly processed, are key components of precision management tools. These tools are often
used to guide variable rate application of inputs (seed, fertilizer, and fungicides). Cotton
producers need tools to aid in‐season management decisions. Scientists have responded to
these needs by developing crop simulation models such as GOSSYM, DSSAT, and many others.
These models require accurate information on soil chemical and physical properties, a detailed
list of management inputs (planting date, seeding rate, fertilizer, irrigation, growth regulator)
and daily weather data including solar radiation, air temperature, rainfall, relative humidity, wind
speed, and wind direction. In this research we propose to use the cotton plant as a probe (or
sensor) to integrate genetic, environmental, and management factors. Plants express the effects
of these interactions in terms of growth, development, health, yield, and fiber quality. Here we
take advantage of the “Digital Twins” concept by using UAS (or satellite) data from any stage of
crop development to create a digital twin model capable of forecasting crop development status
for the rest of the cropping season. Validation of the model shows that the accuracy of
prediction depends on (1) amount of data available as input to the model and (2) length of the
forecasting period. Preliminary results show that a digital twin model created with 30‐40 days of
crop history data can forecast plant height, canopy cover, and biomass for the following 30‐60
days with an accuracy of 93‐97%. The proposed model can be a valuable tool for determining
time and amount of irrigation, growth regulators, fertilizer, and harvest aid application.
properly processed, are key components of precision management tools. These tools are often
used to guide variable rate application of inputs (seed, fertilizer, and fungicides). Cotton
producers need tools to aid in‐season management decisions. Scientists have responded to
these needs by developing crop simulation models such as GOSSYM, DSSAT, and many others.
These models require accurate information on soil chemical and physical properties, a detailed
list of management inputs (planting date, seeding rate, fertilizer, irrigation, growth regulator)
and daily weather data including solar radiation, air temperature, rainfall, relative humidity, wind
speed, and wind direction. In this research we propose to use the cotton plant as a probe (or
sensor) to integrate genetic, environmental, and management factors. Plants express the effects
of these interactions in terms of growth, development, health, yield, and fiber quality. Here we
take advantage of the “Digital Twins” concept by using UAS (or satellite) data from any stage of
crop development to create a digital twin model capable of forecasting crop development status
for the rest of the cropping season. Validation of the model shows that the accuracy of
prediction depends on (1) amount of data available as input to the model and (2) length of the
forecasting period. Preliminary results show that a digital twin model created with 30‐40 days of
crop history data can forecast plant height, canopy cover, and biomass for the following 30‐60
days with an accuracy of 93‐97%. The proposed model can be a valuable tool for determining
time and amount of irrigation, growth regulators, fertilizer, and harvest aid application.