Introduction to Precision Agriculture :
Sources of Soil and Crop Yield Variation within a Field
The objectives of this guide are to illustrate the effect of natural and management induced sources of variation in the field on the spatial patterns seen in yield maps, in field maps of soil parameters and to present interpretation strategies.
Precision Agriculture (PA) is a crop management system that attempts to match the inputs with the actual crop needs for small areas or zones within a farm field. Instead of managing whole fields as a single unit, the management is altered to meet the crop needs of different zones within fields. For PA to be viable, both economic and environmental benefits must be considered, as well as the practical questions of suitable technology such as variable-rate fertilizer spreaders. The management challenge is to optimally manage the zones within the field that have different production capacities. Therefore this means that there may not be a uniform yield throughout the field, rather the yield potential of each zone is maximized.
Field Variability and Remedial Measures
Knowing that yield variation exists is one thing, but deciding what to do about it is another. Crop yields will vary within a field because of available moisture (depends on soil texture, soil depth, organic matter), available nutrients (depends on soil texture, level of nutrients, soil depth), drainage, soil pH, weed pressure etc (Figure 1 ). Many other factors will affect the yield either alone or in combination. Yield maps simply quantify the degree of the variation from point to point, and present it in a visual format. There are many factors that limit crop yields and a yield map only shows that there is a problem but not which factors are limiting the yield. Therefore, interpreting a yield map to make management decisions can be very difficult. The yield map indicates where the problem zones are in the field and the general magnitude of the yield differences. The producer has to determine the causes of the yield variations and whether these factors can be altered by changes in production practices (not all changes are possible or economically feasible). A yield map is a visual report on what happened on the field last year. What happened last year might not happen in the following year either due to a change in the weather pattern or due to the response of a different crop.
Variability in a field is a result of both natural factors and management practices. The soil varies due to soil forming factors such as parent material (marine heavy clay, river sand etc), topography (level or rolling), biological activity (originally a forest -conifers or deciduous-, swamp or grassland), climate (temperature and precipitation) and time. Management practices such as tillage, fertilization and manure spreading can cause variation in the field. Moldboard plows tend to remove soil from the tops of knolls and deposit it downslope - this causes organic matter and nutrients to accumulate downslope. Events from the past, 30 years or more, such as old drainage ditches that are now filled in, can often be distinguished when soil nutrient levels are mapped as shown in Figure 2.
The question is whether the crop yield response to input such as nitrogen (N) or phosphorus (P) varies significantly within a field. If the answer is yes, then under uniform fertilizer or manure application, some parts of the field may not be receiving sufficient levels of nutrients to reach their maximum potential while other parts may be receiving excess amounts, which may have environmental implications. The best situation would be to vary the inputs spatially to meet the varying needs of the crop at different locations. If soil and water systems seem uniform (no change in soil colour, flat topography, etc.) throughout the field and the yield maps show relatively consistent yields (the range between maximum and minimum yields is small) then it may not be economical to apply precision farming technologies such as variable rate applications of fertilizer or lime. In some cases there may be no economically viable method of "fixing" the problem that limits yield such as shallow topsoil depth.
Precision Agriculture: The Components
The application of PA can be seen as a management tool. The choice of agricultural management practices at the field scale that maintain productivity and protect the environment is often a complex procedure. One question is how much information is needed to use PA. Data collection such as soil sampling and interpretation of soil analyses and yield maps can be expensive and time consuming. It is very important to know how this information can benefit crop production and over all decision-making. Some fields require little information to determine the cause of yield variability (rolling topography) while other fields require extensive data collection and even then yield variation may still be unpredictable. Management practices that can fix the cause of reduced yields range from easy to difficult, cheap to expensive and have varying levels of success. The decision to alter the management practices in low yielding areas must be based on the degree and magnitude of yield variability, the correct identification of the reason for the low yield and the economic viability of the remedial actions.
2. Global Positioning System
GPS (Global Positioning System) is used by yield monitors on combines to produce geo-referenced yield maps. Crop yield data is collected every 10 seconds (or other time intervals) over the area determined by the swath width and distance traveled by the combine. Each yield data point has a latitude and longitude position. Each yield point on the map can be directly related to a point in the field. Geo-referencing is the use of latitude and longitude readings from a GPS receiver to place soil nutrient levels, texture, elevation features and yield at their exact location on the field. GPS records are the foundation for site-specific management (precision agriculture). (see Appendix I for more details)
3. Yield Monitor Data
Yield measurement is subject to many random and systematic errors such as:
Many of the yield mapping software associated with the yield monitor will clean-up the data, such as the removal of extremely high or low yield data points. However errors in the yield can still remain on the maps. The producer should carefully examine and evaluate the yield maps using their own knowledge of the field. Yield maps are only an approximation of crop yield at a given location due to several reasons, one is that the flow of grain in the combine does not start and stop abruptly. The crop cut at the edges of the combine header takes longer to reach the sensor than crop cut at the center of the combine header. As well, there is a 'smoothing' effect of the combine, yield monitors tend to overestimate the low yielding areas and underestimate the high yielding areas. The following WebSite has further details on yield monitors http://muextension.missouri.edu/xplor/envqual/wq0451.htm
4. Yield Maps
Yield maps are becoming commonplace however interpreting these yield maps are proving more difficult than producers and agronomes had anticipated - many factors interact to affect crop yield within a given field and year. As well yield maps may not be stable from year to year. The high and low zones may change, depending on climate or other factors. For example in a dry year, the sand soils in a field will have low yields, yet in a wet year these same soils will probably have higher yields than the clays in the same field. Points to remember when evaluating the yield maps.
Yield mapping will be valuable only when producers can turn this information into better management decisions in their operation. However yield mapping is relatively low-cost as compared to intensive soil sampling and it provides complete coverage of the field, which is impossible even when taking soil samples on an intensive grid.
5. Soil Sampling
Field soil samples can be taken using several different methods. The standard method of soil sampling is for each field to be represented by a single composite soil sample which is a mix of several (5 to 10) sub-samples taken throughout the field (generally a single composite sample can represent a maximum size of 4 ha). These sub-samples must represent "average" conditions in the field, that is, the sub-samples should not be taken at field edges, in small depressions, where fertilizer was stacked, where manure was piled or in any other small unrepresentative zone. Sub-samples taken in these locations can have extremely high or low levels of soil nutrients and thus do not represent the average conditions of the field. An un-representative sub-sample will affect the nutrient levels of the single composite sample and therefore change the fertilizer recommendations. Composite soil samples that represent entire individual fields are not geo-referenced.
Geo-referenced soil samples that are used for PA can be taken in one of three ways. By using systematic soil sampling the samples are taken on an intensive grid, for example 40 by 40-m, over the whole field. For many commercial firms, the standard sampling density is one composite sample per hectare, but not on a true grid. The final method uses directed sampling in which soil samples are taken in zones within the fields that are selected by using aerial photos, topography or yield maps. In these three methods the soil samples are geo-referenced. This implies for example that the soil P level has an exact latitude and longitude position in the field. Geo-referenced soil samples are required so that the maps of nutrient levels can be used together with the yield maps to help explain variability in the yield.