Scheduling Methods For
Site-Specific Irrigation

Researchers evaluate Irrigator Pro as a tool for variable-rate irrigation using a site-specific center pivot.

Variable-rate irrigation can provide substantial water savings. However, irrigation scheduling methodologies are needed to precision-apply water for maximum agronomic and economic effectiveness.

Previous research reported that using soil water sensors and tensiometers were superior to using evaporation pan and evapotranspiration (ET) models for scheduling spatial irrigation applications in cotton because the latter two did not take into account soil spatial variability.

Site-specific monitoring of soil water is expensive and time consuming, especially if a field is highly variable and would need many sensors. Irrigator Pro is an expert system designed to manage irrigation decisions based crop variety, previous crop, soil texture, water-holding capacity of the soil, expected yield potential, growing region, crop planting date, current days after planting, rain, irrigation, maximum and minimum soil temperature at 5-cm depth and soil map unit.

An experiment was conducted to evaluate Irrigator Pro as a tool for variable-rate irrigation of peanut using a site-specific center pivot irrigation system.

Irrigator-Pro Vs. Tensiometers
The experiment was conducted in 2007 at the USDA-ARS Coastal Plains Soil, Water and Plant Research Center near Florence, S.C. The study was conducted under a site-specific center pivot system constructed in 1995. Treatments in the study were irrigation based on Irrigator Pro for the predominant soil in a plot (Model-Full Plot), irrigation based on Irrigator Pro for individual soils within a plot (Model-by-Soil), irrigation based on tensiometers and rainfed.

Peanuts (cultivar NC-11) were planted on May 18, in 39-inch wide rows. Clemson University recommendations were used for managing the crop including soil fertility, pest and disease management, and digging and harvesting.

Soil map units in the field were Bonneau loamy fine sand (BnA), Dunbar loamy fine sand (Dn), Dunbar loamy fine sand, overwash (Do), Emporia fine sandy loam (ErA), Goldsboro loamy fine sand (GoA), Noboco loamy fine sand, moderately thick surface (NbA), Noboco fine sandy loam, thick surface (NcA), Noboco fine sandy loam (NfA), Norfolk loamy fine sand, moderately thick surface (NkA), Norfolk loamy fine sand, thick surface (NoA) and Norfolk fine sandy loam (NrA).

Shortly after planting, tensiometers were inserted to a depth of 11 inches (30 cm) directly in a row within each soil map unit within each plot. Soil tension data were collected three times per week and irrigation applications, either one inch or one-half inch depending on level of stress and time of week, were made to individual soils within the tensiometer treatment when soils were at -0.3 MPa.

Soil thermometers were placed in plots managed by Irrigator Pro. These were placed in the predominant soil map unit for the Model-Full Plot irrigation treatment and in each soil map unit in the Model-by-Soil irrigation treatment. The model was run two to three times per week depending on rainfall, and irrigation of the one inch or one-half inch was applied when prescribed by the model.

Irrigator Pro provides recommendations for only three distinct soil categories. The model was run using medium/heavy soil for the Norfolk series and as sandy for the other soils in the experiment.

Within-season measures of canopy temperature and normalized difference vegetative index (NDVI) were made frequently. A global positioning system unit was also mounted on the tractor to allow for the data to be geo-referenced.

Peanuts were dug on Oct. 10, and harvested with a combine one week later.

The 2007 growing season had good early season rainfall followed by a prolonged dry period. A little more than five inches of rainfall occurred during the first eight weeks after planting, but only about two inches of rainfall occurred throughout the rest of the 21-week season.

Irrigator Pro Calls For More Water
The Irrigator Pro model prescribed irrigation beginning between seven and eight weeks after planting for all soils. Soil tensiometers, on the other hand, did not prescribe irrigation until between nine and 10 weeks after planting.

With irrigation applications beginning earlier in the season in plots managed with Irrigator Pro, the plots were generally prescribed more total water for the entire season than application prescriptions for plots in the tensiometer treatment. Total irrigation prescribed for the peanuts in the tensiometer treatment ranged from about seven inches to 14 inches. Eight of 11 soils application prescriptions for the tensiometer treatments were between about 9 and 11 inches.

Irrigator Pro prescribed irrigation ranging from 9 inches to 14 inches with most, five of eight soils, being about 14 inches in the Model-by-Soil treatment and the Model-Full Plot treatment.

Similar Results For All Methods
The three irrigation scheduling methods (tensiometers, Model-Full Plot and Model-by-Soil) did not differ for NDVI at any time during the season. By 16 weeks after planting, there was complete canopy closure and very little variability for this measurement in any of the peanuts that were irrigated.

Mean NDVI was lower and variability for NDVI was greater in the rainfed plots than the three irrigated plots at 14 and 16 weeks after planting. These measurements were taken mid-day, and the severe dry weather caused considerable leaf wilting and leaflet folding in the rainfed plots, which is probably the reason for the lower NDVI values and the higher variability at 14 and 16 weeks after planting than at 10 weeks after planting.

As with NDVI, there was no significant difference among the three irrigation scheduling methods for canopy temperature at either date. Also, variation among soils for the three methods was small at both dates. For the rainfed peanuts, variation among soils was quite large on Aug. 10. By Aug. 21, the drought had intensified substantially, and there was little difference among soils in canopy temperature in the rainfed plots.

Crop yields followed the same trends as the NDVI and canopy temperature data. Rainfed yields were approximately one-half of irrigated yields, but there was no difference among irrigation scheduling methods for yield. Variation for yield was somewhat higher with the tensiometer method than with either method that employed Irrigator Pro.

Use Predominant Soil In Zone
In this study, irrigation by any scheduling method increased yield. Irrigation also increased NDVI at sampling dates later in the season. Variability among the different soil map units for canopy temperature was lower for all irrigated treatments than for the rainfed treatment at the beginning of a long rain-free period, but not when water deficit stress was severe later in the season.

Using the predominant soil map unit within an irrigation management area was just as effective as irrigating on a by-soil basis. Both treatments that used the model prescribed more irrigation water than the tensiometer method.

The research is continuing on evaluating the expert system in the Carolina’s growing region. PG

Information provided by Philip Bauer, Kenneth Stone, Warren Busscher, Joseph Millen, Dean Evans and Ernest Strickland, USDA-ARS, Coastal Plains Soil, Water and Plant Research Center Florence, S.C.