Computer weather model helps farm decision-making

Given their druthers, most farmers would prefer to have their backsides perched comfortably (or not) on the foam padding of a tractor seat rather than in a butt-wearing office chair in front of a blinking screen.

But progress, as they say, marches on and decisions farmers make at their computers often mean a more satisfying ride as they plow, cultivate and harvest. Texas A&M research engineer Carlos Fernandez, who plies his trade out of the A&M Research and Extension Center in Corpus Christi, is trying to make that office chair a little more comfortable and that computer screen a bit less daunting, so growers will use historical and real-time data to help determine what and when to plant, and when to irrigate, spray, defoliate and harvest.

Fernandez has been building a program designed to use weather data to help farmers make critical crop decisions. The program depends on data gathered from 20 weather stations scattered throughout the Coastal Bend and the Upper Coastal Bend areas of South Texas.

Fifteen of these weather stations once belonged to the Weather Station Network Program developed in the early 1990s by Juan Landivar but discontinued in 1999 due to incompatibilities with Windows updates as well as Y2K compliance issues. Fernandez took over in 1999 and built “The Crop-Weather Program for South Texas,” taking advantage of the increased ability of computers and the Internet to deliver information.

“We needed to either upgrade Landivar's program through a number of software generations or build an entirely new program taking advantage of the Internet, which was just taking off,” Fernandez says. He added weather stations and designed a program that “no longer must be installed onto a user's computer. It's available via the Web, which removes a burden from the user's side.”

Ease of use

Making the program easy to use takes precedence for Fernandez. Currently, he employs as many drop-down menus as possible to reduce typing chores. That also eliminates mistakes.

“Farmers are intelligent,” Fernandez says, “But most don't have time to master a complex computer program. We have to make this program simple. Providing information through the Internet helps us do that.” He's also added more data farmers can use for decisions.

“Historically, farmers make management decisions based on experience and instinct,” Fernandez says. He admits that no program can take all the guesswork out of production agriculture, but data collected from weather stations, combined with a specific farmer's field data, crop history and resource list provide at least a template to begin the process. And real-time or as current as possible weather data, available on a computer screen, may allow farmers to make split-second decisions on irrigation, pesticide applications and harvest.

“Farmers get temperature, relative humidity, solar radiation, soil temperature at three depths, rainfall amounts, and wind speed and direction, so they have a lot of data to help with day-to-day decisions,” Fernandez says.

The program has “been developed through the eyes of the end-user, who doesn't need a tremendous amount of computer skill to use it.”

Develop baseline

They simply access weather station data gathered over the Internet. The critical aspect, and the facet that could make the program one of the most useful management tools on a farm, is the data farmers enter. They can develop a baseline for specific fields, indicating soil types, crop history, elevation, etc., and build layers of data onto that base with subsequent years' information.

Fernandez says farmers can identify management units — farms, fields, even parts of fields — and use program data to fine-tune management on those units.

“After registering, growers can use a suite of tools to apply to each unit. They need to develop a field profile for each management unit and, since this information is kept in a database, they do this only once.” Profile data include variety, soil type, row width, planting date, planting depth, and planting rate.

Field profiles allow farmers to watch crop progress throughout the season and take appropriate actions. Soil temperature data, for instance, help determine best planting dates. “We can offer a 10-day view so growers can determine trends.”

After planting, growers keep up with daily minimum temperatures at seed depth to determine if damaging low temperatures occur during germination.

The system also keeps track of soil moisture and calculates, hour-by-hour, the “water balance throughout the soil profile. This helps determine what crops or which varieties to plant,” Fernandez says.

Keep water tabs

“We can also see where in the profile water is located. If growers run a soil water balance calculation they can keep tabs on it like they would a checkbook.”

Those calculations, including water loss and rainfall or irrigation accumulations, provide a template for irrigation timing, Fernandez says.

Keeping track of degree-days for cotton gives growers a better feel for when to defoliate and to prepare for harvest.

Fernandez says data also may show farmers how crops stand up to pest infestations.

Currently, the program is designed for cotton but Fernandez is tweaking it to fit other crops.

Eventually, the project will include a “crop accountant” program that includes budgets, on-farm resources such as equipment, production costs, labor and other items. It may use weather data to project yield potential.

“This is an evolving project,” Fernandez says. “Its modular design has reusable components and allows easy integration of other developers. It is also scalable so we can expand the user base. We would like to make this program usable in production areas beyond the Coastal Bend and Upper Coast, something that we will be working on if we procure additional funding for this project.

“We think the program has a lot of potential,” Fernandez says.

The Crop-Weather Program for South Texas can be accessed at its dedicated Web site

e-mail: [email protected]

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.