Datasets:

text
stringlengths
0
1.09k
Increased sprinkler application uniformity will often result in increased yields, decreased runoff, and decreased percolation.
Improved sprinkler uniformity can be desirable from both economic and environmental standpoints.
Their study indicated irrigation nonuniformity can result in nutrient leaching from over-irrigation and water stress from under-irrigation.
Both problems can cause significant economic reductions.
Sprinkler irrigation does not necessarily have to be a uniform broadcast application to result in each plant having equal opportunity to the irrigation water.
Equal opportunity can still be ensured using a LEPA nozzle in the furrow between adjacent pairs of crop rows provided runoff is controlled.
Figure 4.
LEPA concept of equal opportunity of plants to applied water.
LEPA heads are centered between adjacent pairs of corn rows.
Using a 5-ft nozzle spacing with 30-inch spaced crop rows planted circularly results in plants being approximately 15 inches from the nearest sprinkler.
After Lamm.
Some sprinkler application nonuniformity can also be tolerated when the crop has an intensive root system.
When the crop has an extensive root system, the effective uniformity experienced by the crop can be high even though the actual resulting irrigation system uniformity within the soil may be quite low.
Additionally, when irrigation is deficit or limited, a lower value of application uniformity can be acceptable in some cases as long as the crop economic yield threshold is met.
Many irrigators in the U.S.
Great Plains are using wider in-canopy sprinkler spacings in an attempt to reduce investment costs.
Surveys from western Kansas in 2005 and 2006 indicated only 34% of all sprinkler systems with nozzle height of less than 4 ft had consistent nozzle spacing less than 8 ft.
Sprinkler nozzles operating within a fully developed corn canopy experience considerable pattern distortion and the uniformity is severely reduced as nozzle spacing increases.
Figure 5.
Differences in application amounts and application patterns as affected by sprinkler nozzle height and spacing.
Center pivot sprinkler lateral is traversing parallel to the circular corn rows.
Data are from a fully developed corn canopy, July 1996, KSU Northwest Research-Extension Center, Colby, Kansas.
Data are mirrored about the nozzle centerline for display purposes.
Arrows on X-axis represent location of corn rows and thus the location for higher stemflow amounts.
Although Figure 5 indicates large application nonuniformity, these differences may or may not always result in crop yield differences.
Hart concluded from computer simulations that differences in irrigation water distribution occurring over a distance of approximately 3 ft were probably of little overall consequence and would be evened out through soil water redistribution.
Some irrigators in the Central Great Plains contend that their low capacity systems on nearly level fields restrict runoff to the general area of application.
However, nearly every field has small changes in land slope and field depressions which do cause field runoff, in-field redistribution or deep percolation in ponded areas when the irrigation application rate exceeds the soil infiltration rate.
In the extreme drought years of 2000 to 2003 that occurred in the U.S Central Great Plains, even small amounts of surface water movement affected sprinkler-irrigated corn production.
Similarly some of the worst erraticity in sprinkler-irrigated corn observed in the summer of 2011 was for sprinklers with 10 ft spaced in-canopy sprinkler packages.
Figure 6.
Large differences in corn plant height and ear size for in-canopy sprinkler application over a short 10-ft.
distance as caused by small field microrelief differences and the resulting surface water movement during an extreme drought year, Colby, Kansas, 2002.
The upper stalk and leaves have been removed to emphasize the ear height and size differences.
Figure 7.
Erraticity of sprinkler irrigated corn in southwest Kansas in 2011 under extreme drought conditions thought to be related to a nozzle spacing too wide for incanopy application.
CROP ROW ORIENTATION WITH RESPECT TO DIRECTION OF SPRINKLER TRAVEL
When using in-canopy sprinkler application, it has been recommended that crop rows be planted circularly so that the crop rows are always perpendicular to the center pivot sprinkler lateral.
Matching the direction of sprinkler travel to the row orientation satisfies the important LEPA Principles 2 and 5 noted by Lyle concerning water delivery to one individual crop furrow and equal opportunity to water by for all plants.
Producers are often reluctant to plant row crops in circular rows because of the cultivation and harvesting difficulties of narrow or wide "guess" rows.
However, using in-canopy application for center pivot sprinkler systems in non-circular crop rows can pose two additional problems.
In cases where the CP lateral is perpendicular to the crop rows and the sprinkler spacing exceeds twice the crop row spacing, there will be nonuniform water distribution because of pattern distortion.
When the CP lateral is parallel to the crop rows there may be excessive runoff due to the great amount of water being applied in just one or a few crop furrows.
There can be great differences in incanopy application amounts and patterns between the two crop row orientations.
Figure 8.
Two problematic orientations for in-canopy sprinklers when crops are not planted in circular rows.
Figure 9.
Differences in application amounts and application patterns as affected by corn row orientation with respect to the center pivot sprinkler lateral travel direction.
Dotted lines indicate location of corn rows and stemflow measurements.
Data are from a fully developed corn canopy, July 23-24, 1998, KSU Northwest Research-Extension Center, Colby, KS.
Data are mirrored about the centerline of the nozzle.
PATTERN DISTORTION AND TIME OF SEASON
Drop spray nozzles just below the center pivot sprinkler lateral truss rods have been used for over 30 years in northwest Kansas.
This configuration rarely has had negative effects on corn yields although the irrigation pattern is distorted after corn tasseling.
The reasons are that there is only a small amount of pattern distortion by the smaller upper leaves and tassels and this distortion only occurs during the last 30 to 40 days of the irrigation season.
In essence, the irrigation season ends before a severe soil water deficit occurs.
Compare this situation with spray heads at a height of 1 to 2 ft that may experience pattern distortion for more than 60 days of the irrigation season.
Under dry and elevated evapotranspiration conditions in 1996, row-to-row corn height differences developed rapidly for 10-ft spaced sprinkler nozzles at a 4 ft nozzle height following a single one-inch irrigation event at the KSU Northwest Research-Extension Center, Colby Kansas.
A long term study at the same location on a deep silt loam soil found that lowering an acceptably spaced spinner head from 7 ft further into the crop canopy caused significant row-to-row differences in corn yields.
Figure 10.
Crop height difference that developed rapidly under a widely spaced in-canopy sprinkler following a single 1 inch irrigation event at the KSU Northwest Research-Extension Center, Colby, Kansas.
Photo taken on July 6, 1996.
Figure 11.
Row-to-row variations in corn yields as affected by sprinkler height for 10 ft.
spaced in-canopy sprinklers.
Sprinkler lateral travel direction was parallel to crop rows.
Data was averaged from four irrigation levels for 1996 to 2001, KSU Northwest ResearchExtension Center, Colby, Kansas.
COMBINATION OF EFFECTS CAN CAUSE ERRATICITY
Sometimes poor design, installation or maintenance problems can exist for years before they are visually observed as sprinkler irrigation erraticity.
It may take severe drought conditions for some of these subtle effects to combine to such an extent to be noticeable erraticity.
In addition, smaller row-to-row differences in crop yield cannot be measured with yield monitors on commercial-sized harvesters.
An example of a combination several of these subtle effects was observed during the severe drought of 2002 in northwest Kansas.
The small nozzle height difference on this sprinkler allowed at least three small effects to combine negatively to cause the sprinkler erraticity:
1.
Since there are no pressure regulators, the small height difference results in unequal flow rates for these low pressure spray nozzles.
2.
There is a incorrect overlap of the sprinkler pattern due to the height difference with one sprinkler within the canopy while the other two nozzles are above the canopy.
3.
Evaporative losses would be greater for the nozzles above the crop canopy.
Figure 12.
Erraticity of sprinkler-irrigated corn near Colby, Kansas during the extreme drought year of 2002.
The drought that southwest Kansas experienced in 2011 was devastating to production on many sprinkler irrigated corn fields, but the erraticity did highlight some design and management issues that producer might address before the next irrigation season:
1.
Does the selected sprinkler package strike the correct balance in reducing evaporative losses without increasing irrigation runoff or in-field water redistribution?
2.
Does the sprinkler package and its installation characteristics provide the crop with equal opportunity to applied or infiltrated water?
3.
Are the sprinkler nozzle heights and spacings appropriate for the intended cropping?
4.
Should planting of taller row crops such as corn be in circular patterns if incanopy sprinklers are used?
5.
Are there subtle irrigation system characteristics that might combine negatively to reduce crop yields?
These design and management improvements won't change the weather conditions, but they might change how the crop weathers future droughts.
This paper is the result of cooperative efforts of the authors through the Ogallala Aquifer Program, a consortium between USDA Agricultural Research Service, Kansas State University, Texas AgriLife Research, Texas AgriLife Extension Service, Texas Tech University, and West Texas A&M University.
This is a joint contribution from Kansas State University, USDA Agricultural Research Service and Texas A&M University.
Contribution no.
12-309-A from the Kansas Agricultural Experiment Station.
This paper was first presented at the Central Plains Irrigation Conference, February 21-22, 2012, Colby, Kansas.
It can be cited as
DOES DEFICIT IRRIGATION GIVE MORE CROP PER DROP?