High-tech farming: today the tractor, tomorrow the soil sensors

Successive generations of tractors, harvesters, and other mechanical ``hired hands'' have transformed American agriculture -- boosting the farmer's production, and often his costs, to once unheard-of levels. Farm mechanization has been a revolution of sorts, with men like Roger Garrett, dean of the agriculture engineering department at the University of California, Davis, vigorously at its forefront. He has been on the faculty at Davis, one of this country's premier agricultural schools, for 23 years. A mechanical harvester for lettuce and a mechanical thinner for the same crop are among his invention credits.

Dr. Garrett, a soft-spoken man with an informal manner, took a couple of hours recently to look back on what's been accomplished in his field and what's likely to happen in the future.

He sees a wave of automation on the horizon that could again transform farming. ``We're really on the verge of another revolution in agricultural technology, in our ability to process data and control operations,'' Garrett says. Tomorrow's agricultural marvels, he explained, won't be supercombines or high-tech tractors, but electrical sensing devices that will monitor such things as soil chemistry and quickly tell a farmer when conditions are optimal for fertilizing or planting.

This type of automation won't displace the mechanical behemoths that currently plow, plant, and harvest. Rather, says Garrett, it will ``make those machines capable of sensing the conditions under which they're operating'' -- conditions ranging from soil makeup to climate. The sensing devices will be teamed with existing computer technology to help the farmer arrive at a clear idea of the response those conditions demand.

``With computers we have the ability to receive information from a number of sensors and do rather sophisticated analysis of that information,'' he explains. The central challenge in this automation scenario, in his view, is the development of low-cost, reliable sensing devices that meet the particular needs of the farmer. The sensors now available have been developed from other uses, Garrett says, and need refining by researchers within agriculture.

A further challenge, he adds, is interpreting ``what the data really means.'' As an example, he offers a simple case of drying grain with the traditional means of fans and heaters. One could drop a temperature probe in the drying chamber, he says, ``but just having the temperature doesn't tell us what to do with it.'' The next step, he continues, would be the construction of ``some good models, mathematical models, to assess the function of the operation and the behavior of the grain.''

The upshot of this, he suggests, could be a decision to position the temperature device where the air goes out. A certain level of heat would indicate that less moisture was entering the air and the grain was satisfactorily dry.

A more complex use of sensor technology might measure the ``differing levels of fertility in soil,'' says Garrett. Information about the amount of organic matter in the soil could help the farmer decide when to apply chemical herbicides and how much to apply. The sensor device, he explains, would likely be attached to the machine that applies the herbicide. For some of these herbicides, he notes, ``there's very little threshold between too little and too much.'' Reliable sensing of soil conditions could

reduce costs to the farmer by avoiding overuse of the herbicide and at the same time better protect the environment, he argues.

Looking back on progress in farming, Garrett observes: ``So far, what we have done in agriculture is provide a substantial increase of power -- machines to replace humans and animals.'' He points out that mechanization has gone hand in hand with other advances in farming -- new breeds of vegetables that stand up to mechanical harvesting, for instance, or new pesticides and fertilizers that need the precision of mechanical spreading to be most effective. Farming, in this sense, is an ``integrated system. ''

It's integrated with society at large, too, notes Garrett. Advances in farm machinery have helped formerly marginal land become productive. And that, Garrett says with a bit of sadness, has ``allowed the country to sacrifice a lot of good farmland that went to urban sprawl.''

Mechanization has also meant competition with traditional human manpower on farms. A lawsuit against the University of California, charging that its mechanization research has worked to the detriment of family farms and displaced workers, is pending in state court. While there still are protests on this issue, says Garrett, they are minimized at present. He says the push toward mechanical harvesting of many of California's vegetable and fruit crops has trailed off, both because planters have arrived at some agreement with labor and because of ``the readily available field labor we have here.''

Like anyone in agriculture, Garrett is sharply aware of the ironies nurtured by advances in farm technology. Commenting on the overproduction problems that stalk much of American farming, he observes that ``agriculture's biggest problem is its success.''

``For the last 30 years we've been saying that a food crisis is coming. We've tried to stay a step ahead of that, and we have -- more than a step.''