The value of remote sensing for weed monitoring depends on the crop

Feb 12, 2020

At the recent CWSS meeting in Monterey there were a number of presentations on the impact of precision and robotic technology in weed management. The weed school consisted of presentations that detailed technology for large scale monitoring of agricultural fields as well as small autonomous weeding robots that differentiate crops from weeds and selectively kill the weeds in the same pass. Both scales of technology have their place. Large-scale cropping systems such as wheat in the great plains are grown on very large farms where efficiency and ability to monitor large areas at low-cost is essential. California vegetable farms tend to be much smaller and more intensive with workers in every field most every day for harvesting, irrigation, fertility and pest management activities such as scouting and spraying. On large scale farms in the great plains covering thousands of acres, it is impossible for growers to cover such a large area with great frequency.

This piece compares the value of aerial monitoring of weeds in both systems. For a large wheat farm knowledge of the location of a severe weed patch such as Canada thistle would allow a grower using remote sensing data to target the areas of the farm infested with this weed. This would allow efficient use of the grower's time and resources by targeting the application to the area of the farm where needed such as an infestation of Canada thistle. Wheat has postemergence rescue options for Canada thistle such as clopyralid and so the aerial monitoring probably adds value. However, aerial monitoring of a crop like lettuce for weeds would provide little value. An aerial map of a lettuce field weeds would not be useful for targeted spraying as lettuce has no good herbicide options for rescue treatments. The only rescue treatment for lettuce weeds is mechanical cultivation and hand weeding – neither would benefit from weed mapping on the scale of a 10 to 15 acre planting.

The Canada thistle example is a two-step process, detection with the aerial monitoring, and then weed control with the targeted spray application. To target the spray application, the aerial images gathered must be interpreted so that the spray can be targeted to the proper location. Growers may be too busy or lack the skills needed to interpret data, and are forced to pay an expert to interpret the data and tell them where to spray. Where the value of the information exceeds the cost then remote sensing is worthwhile. Automated weeders detect and control the weeds in the same pass in vegetable fields. There is no need for the aerial imaging and interpretation step with these automated weeders. Extensive wheat growers need rapid and inexpensive monitoring of the weeds to target rescue sprays. There are no rescue herbicides for most vegetables. It is important to make this distinction as discussions of agricultural technology, at least at the national level, are dominated by the broad acre agronomic crops. Agronomic crops are a larger market than vegetables and so the national equipment manufacturers naturally focus on the large crops. California has highly segmented production systems consisting of hundreds of crops, and the technology needs are very different. Drones have little value for weed monitoring in vegetable and strawberry plantings due to lack of rescue herbicides. If drones can detect insects and diseases, then possibly they may have more value in these valuable crops. The simple fact remains – taking pictures of weeds does not control them. Bottom line – producers of intensive horticultural crops should make sure that a remote sensing service provider is worth the cost.

By Steven Fennimore
Author - CE Weed Specialist