What’s Precise About Precision Agriculture? May 2001

R.L. (Bob) Nielsen
Purdue Agronomy
rnielsen@purdue.edu

Precision Agriculture refers to the use of GPS-enabled technologies for managing crops on a site-specific basis within fields. Some of the necessary tools available for site-specific crop management (SSCM) are here and/or are being developed rapidly. These tools include grain yield monitors, intensive soil-sampling procedures, variable rate controllers for crop inputs, remotely sensed imagery, portable GPS-enabled data recorders and the software to collect and ‘massage’ the georeferenced data.

The availability of these tools or services has generated a lot of interest and enthusiasm among early adopters of the technology. Growers who adopt the technology in some form or another do so with the expectation that SSCM will enable them to a) improve grain yield and/or minimize the variability of yield within fields and/or b) lower their per unit production costs.

One of the realities that adopters of the technology need to understand is that crops are influenced by a vast array of yield influencing factors (YIFs). Some of these YIFs influence yield directly (e.g., phytophthera on soybean). Some influence yields by interacting with other YIFs (e.g., poorly drained soils + phytophthera on soybean). Some YIFs occur every year unless modified by crop management strategies (e.g., low soil pH). Other YIFs do not reoccur every year (e.g., drought). Some YIFs influence crops differently (e.g., soybean cyst nematode on soybean versus corn). The one common denominator is that weather interacts with almost all YIFs in determining their severity of influence on crop grain yield.

Those YIFs that occur every year AND in the same spatial pattern within the field can be identified and potentially managed with SSCM strategies. I call such YIFs ‘perennial’ in nature. Two of the most often cited examples of improved profits from SSCM are soil pH and soil drainage patterns, which happen to also be good examples of ‘perennial’ YIFs.

The problem is that many YIFs are not ‘perennial’, but are more sporadic in their occurrence both spatially (within field) and temporally (over years). Disease and insect pressures are good examples of ‘sporadic’ YIFs. Site-specific technology can identify these problems when they occur, but not necessarily prevent their recurrence because of their unpredictability (often influenced by weather patterns).

The bottom line in the adoption of SSCM strategies is that SSCM itself requires season-long crop monitoring that is also site-specific. In other words, interpretation of yield maps will require many other maps or layers representing various YIFs that were also measured and recorded on a georeferenced basis within a field. Such crop scouting will require the services of competent agronomists with broad backgrounds in agronomy, entomology, plant pathology and weed science. Of course, georeferenced crop scouting will also require the investment in GPS-enabled equipment and software necessary for georeferenced crop scouting.

A weak link in the SSCM scenario is that the ability to integrate layers of spatial agronomic information and relate them to spatial yield variability requires a working knowledge of geostatistics that most practicing agronomists simply do not yet have. Additionally, the availability of low-cost user-friendly GIS software programs is very minimal. Higher-priced GIS software programs capable of performing some of these geostatistical analyses are often not user-friendly and require a solid understanding of geostatistics. Consequently, farmers and agronomists often look for relationships between layers of cropping information with their eyeballs and erroneous correlations are often made.

For example, you watch the grain yields drop on the monitor as you harvest through that thick patch of foxtail. The natural correlation to be made is that the foxtail infestation resulted in severe yield loss. However, the real answer may lie in the possibility that the only reason the foxtail were present was that extensive ponding earlier in the season killed off most of the crop in that area and the weeds simply took advantage of the open space.

Another factor to be aware of when working with spatial data is that GIS mapping programs create those pretty color maps by interpolating values between all the known values of a spatial data set. Spatial data sets may be either dense (many data points per acre) or sparse (few data points per acre). A good example of a dense data set is a yield data file. A good example of a sparse data set is that resulting from a 2-½ acre soil-sampling scheme.

Dense data sets have fewer ‘holes’ per holes than do sparse. Consequently, GIS mapping programs require less interpolation to create maps from dense data sets than from sparse data sets. Consequently, the maps resulting from the interpolation procedure on dense data sets are inherently more believable. Conversely, maps created by interpolating sparse data sets must be taken with the proverbial ‘grain of salt’. Because of this caution, one should always sample YIFs as densely as time and money will allow. From the perspective of crop scouting or monitoring, one can never have too much data.

Because of some of these current limitations with the technology, it may be that we can only fine-tune our crop production to a limited extent with SSCM strategies. We can identify and correct certain obvious YIFs such as soil fertility, pH, and drainage. We will be much more limited in our ability to manage those YIFs that occur more sporadically.

The good news is that growers who begin adopting SSCM strategies often spend more time thinking about their farming operation than before. They develop an increased awareness and understanding of their crops, weeds, diseases, insects, weeds, soils, etc. One benefit to this is that they often are better equipped to identify crop problems early in the season, often allowing for more effective problem management. Similarly, SSCM strategies can help direct growers’ attention to areas of fields with special peculiarities that may have missed their attention before.