Center Pivot Irrigation System Evaluation
Center Pivot Irrigation System Evaluation
In Michigan
March 2011
1
Table of Contents
Introduction ..................................................................................................................................... 5
Background and Recent Research ................................................................................................... 7
Objectives ........................................................................................................................................ 8
Study Area and Systems Description ............................................................................................... 8
Methodology and Equations ........................................................................................................... 9
Results and Discussions ................................................................................................................. 12
Fertigation based on Uniformity ................................................................................................... 18
Recommendations......................................................................................................................... 18
Acknowledgements ....................................................................................................................... 23
References ..................................................................................................................................... 24
Appendix A .................................................................................................................................... 25
Appendix B..................................................................................................................................... 30
Appendix C..................................................................................................................................... 36
2
List of Tables
List of Figures
Figure 1. Area covered by each sprinkler increases as the distance from the pivot center
increases. ............................................................................................................................. 6
Figure 2. Map of Tekonsha, Marshall, and Constantine MI. where evaluations tests were
conducted. ........................................................................................................................... 9
Figure 3. Uniformity Coefficient and Distribution Uniformity for all evaluation tests.... 14
Figure 4. Nitrogen distribution based on water applied and sprinkler system uniformity in
test 4. ................................................................................................................................. 19
Figure 5. Nitrogen distribution in test 4 assuming 65 lbs/acre applied over the season by
fertigation in test 4. ........................................................................................................... 20
Figure 6. Graph based on 200 lbs nitrogen application over a season via fertigation in test
4......................................................................................................................................... 20
Figure 7. Nitrogen application was calculated on test 2 with the greatest uniformity rating
(89%) assuming all nitrogen is applied via fertigation (200 lbs. over the season). .......... 21
Figure 8. Rotating Sprays mounted on the lateral of the center pivot. ............................. 25
Figure 9. Catch cups lined up along the edge of the corn crop. ........................................ 25
Figure 10. Measuring water volume from catch cup. ....................................................... 26
Figure 11. Row of cups before irrigation line has passed over. ........................................ 26
Figure 12. Using the Ultra-sonic flow meter to measure flow rate through the system. .. 27
Figure 13. Center pivot with end gun on. ......................................................................... 27
Figure 14. Water leaking from the lateral joint on the center pivot line. .......................... 28
Figure 15. Center pivot irrigation operating. .................................................................... 28
Figure 16. Rotating sprays suspended from the lateral of the center pivot. ...................... 29
Figure 17. End Gun off during the evaluation test............................................................ 29
Figure 18. Water distribution of Test 1. ............................................................................ 30
Figure 19. Water distribution of Test 2. ............................................................................ 30
Figure 20. Water distribution in test 3. ............................................................................. 31
Figure 21. Water distribution in test 4. ............................................................................. 31
Figure 22. Water distribution in test 5. ............................................................................. 32
Figure 23. Water distribution in test 6. ............................................................................. 32
Figure 24. Water distribution in test 7. ............................................................................. 33
Figure 25. Water distribution in test 8. ............................................................................. 33
Figure 26. Water distribution in test 9. ............................................................................. 34
Figure 27. Water distribution in test 10. ........................................................................... 34
3
Figure 28. Distribution of nitrogen when applying 65 lbs/acre nitrogen fertilizer by
irrigation in test 2. ............................................................................................................. 35
Figure 29. Soil texture identification triangle. .................................................................. 36
4
Introduction
Center pivot irrigation systems are invented over 60 years ago to reduce labor
requirements, enhance agricultural production, and optimize water use. According to
USDA Farm and Ranch Irrigation Survey in 2008 (2), center pivot irrigation are used on
the majority of sprinkler-irrigated land in United States and represent 83% from all types
of sprinkler systems.
A center pivot consists of a lateral circulating around a fixed pivot point. The lateral is
supported above the field by a series of A-frame towers, each tower having two driven
wheels at the base.
Water is discharged under pressure from sprinklers or sprayers mounted on the laterals as
it sweeps across the field or suspended by flexible hose over the crops. The lateral line is
rotated slowly around a pivot point at the center of the field by electric motors at each
tower.
Uniformity of a system is a measure of its ability to apply the same depth of water to
every unit area. Without good uniformity, it is impossible to irrigate adequately and
efficiently; parts of the field will be either over-irrigated or under-irrigated.
Three uniformity measurements are to be considered in the evaluation; Coefficient of
Uniformity (CU) and Distribution Uniformity (DU) and Potential Application Efficiency
of Low Quarter (PELQ).
The CU accounts for the increased area covered by each sprinkler as you move further
from the pivot center. Sprinklers near the end gun cover greater acres than those close to
the center pivot (Figure 1).
5
Figure 1. Area covered by each sprinkler increases as the distance from the pivot center
increases.
compares the lowest quarter of the water depth caught to the entire set of data from
the catch cans. is useful as an indicator of the magnitude of the distribution problems.
DU is calculated by dividing the weighted average of the lowest 25% of the catch cans by
the weighted average of the entire catch cans.
A of 85% or greater is considered excellent, 80% is considered very good, 75% is
considered good, 70% is considered fair, and 65% or less is considered poor and
unacceptable (6).
PELQ should be determined in order to evaluate how effectively the system can
utilize the water supply and what the total losses may be. It is, therefore, a measure of
6
the best management practice and should bet thought of as the full potential of the
system.
Most irrigation equipment in Michigan has not been evaluated for system uniformity.
Systems can be new to over 25 years old or older. Older systems can have greater water
losses due to leaking joints, clogged sprinklers, rusted equipment, etc.
Knowledge of changes in the magnitude water applied over time is important to
determine the causes of deficiencies in application rates and uniformities. Non-uniform
water application leads to over or under irrigation in various parts of the field which can
result in wasted water and energy and the potential for nitrogen leaching. This
information is needed to efficiently and effectively manage irrigation.
Water is pumped from a well or nearby water source to the center of the pivot where it is
distributed along the lateral pipe. Water is applied through sprinklers on that can be
attached directly to the pipe or hang down on hoses called drop nozzles.
Many recent developments have focused on improved control of center pivot irrigation
systems and incorporation of GPS equipment to all application of varying depths of water
to different field sectors. Manufactures and researchers are also working to integrate soil
water and plant sensors into center –pivot control. This enhanced monitoring promises to
optimize water use but at a high price for the time being (2).
For more improvement to the system performance, center pivot may be provided with
self-powered infrared thermometers and a GPS receiver on a center pivot lateral,
additional with remote spatial and temporal crop monitoring is accomplished by locating
sensors within a field. The resulting is automatic irrigation scheduling without using any
traditional tools for soil water content sensing and without using the traditional irrigation
scheduling (2).
7
Objectives
The objectives for this study are:
Center pivot systems in Tekonsha and Marshall consist of 7 and 8 towers plus end gun
tower respectively. Rotating sprays are used in Tekonsha field and fixed sprays in
Marshal. The field in Constantine consists of 4 tower plus an end gun with rotating sprays
suspended from the lateral.
Lengths of the towers and numbers of sprinklers are different from one system to the
other.
Sources of irrigation water are ground water, local rivers, and ponds.
Fields in Marshall and Constantine are flat; the field in Tekonsha has a gently rolling
topography.
The evaluation test method and procedure are performed based on the ASAE standard
S436.1 (1), and Merriam and Keller (6).
Soil samples were taken from each evaluation field before the irrigation system was
turned on. Soil type and moisture content was determined at MSU soil laboratories. All
fields’ soil is Sandy Loam except the field in Constantine town was Loamy Sand. Soil
analysis tests are shown in Table 6 (Appendix C).
8
At each test, temperature, wind speed and direction, relative humidity, evaporation losses,
system flow rate, speed setting, wetted radius for the sprinkler and system operating
pressure are recorded. Moreover, crop height, conditions and root depth are also taken.
Uniformity tests were conducted following the ASAE S436.1 standard for center
pivots. Under the standard, catch cups are spaced 3m (10ft) apart in 1 or more rows
extending from the pivot center straight out to the circle edge. When the pivot is started,
no water should be entering the cups until the unit is at full pressure and speed.
Uniformity Coefficient ( ):
Modified Heermann and Hein formula will be used (1) to calculate the Coefficient of
Uniformity ( ) as follows:
Figure 2. Map of Tekonsha, Marshall, and Constantine MI. where evaluations tests were
conducted.
9
∑
(1)
∑
Where:
= Coefficient of Uniformity, (%).
n = number of collectors used in the evaluation.
i = number assigned to identify a particular collector beginning with i=1
and ending with i=n.
= volume or depth of water collected in the ith collector.
= distance of the ith collector from the pivot point.
= weight average of the volume (or depth) of the water caught.
∑
(2)
∑
(3)
Where:
= low quarter Distribution Uniformity, (%).
The average weighted system catch ( is found (4) by dividing the sum of the
weighted catches by the sum of the catch location where cups are placed. For the average
minimum weighted catch ( , an unknown number of cups that represents
the low ¼ of the irrigated area must be used. The low ¼ is selected by picking
progressively larger (unweight) catches and keeping a running total of the associated
location until the subtotal approximately ¼ of the sum of all the catch location. The
average weighted low ¼ of the catch is then found by dividing the sum of the ¼ of the
weighted catches by the sum of the associated catch location
10
Q x T= (4)
Where:
Q = Flow Rate (gpm).
T = Time for full Revolution (min).
= Volume Pumped to System (gallons).
Time of revolution is taken from the farmer or from the information book of the
manufactures, which is an approximate time. Plus or minus fifteen minutes in the
estimation time will affect the total water volume applied through the system by 2%
differences.
The total volume of water caught by the cups is calculated by multiplying the area
covered by each cup times the corresponding water depth.
(5)
Where:
= Area covered by each catch can (acres).
= depth of water caught (in.).
Total Volume caught in Catch Cans (gallons).
All the analyses of the field data are plotted on MSU Excel extension irrigation system
evaluation (8).
Table 1. Specifications for the selected field tests.
Field System
Name System flow Pressure End
Test no. Location Date of test (gpm) (psi) gun Type of crop
1 A Tekonsha June-28-2011 750 70 Yes Corn & Soybean
2 A Tekonsha June-29-2011 750 70 Yes Corn & Soybean
3 B Burlington June-30-2011 650 28 No Corn
4 C Marshall July-05-2011 800 41 No Corn
5 A Tekonsha Aug.-03-2011 619 65 Yes Corn & Soybean
6 D Tekonsha Aug.-11-2011 847 44 Yes Corn
7 C Marshall Sept.-09-2011 788 35 Yes Corn
8 E Constantine Oct.-14-2011 420 31 No Seeds Corn
9 E Constantine Oct. 28-2011 429 NA No Seeds Corn
10 E Constantine Oct.-28-2011 534 56 Yes Seeds Corn
* Test 6 was conducted on a different center pivot system at a different field than in tests
1, 2, and 5.
11
Results and Discussions
Average depth applied, CU and DU are listed in Table 2. Moreover, CU and DU for all
tests are plotted in Figure 2 to show variations and tendencies.
In tests 2 and 5 the CU and DU values are good and excellent, respectively. Tests 4 and
7, the same evaluation test is done under good weather conditions, except in test 7 the end
gun is operated and in test no.4 the end gun is off. Both tests rendered CU and DU values
below the acceptable limits. Reasons for these low values can be attributed to the
following:
1) Water leakage from the system is affecting the performance of the sprinklers and the
distribution of the water pressure at the sprinklers outlets. In test 2, if tower 6 leaks
are fixed (Figure 9), this would increase the coefficient of uniformity by 1%. This
change leads to a significant increase because the can is far from the center pivot.
The sprinklers cover more area by acreage as they move farther from the center, thus
carrying more weight when calculating the . This change was also applied to test 5
of the same field, with a leak at tower 6, and resulted in a 2% increase in .
3) Saving energy, water and operation cost are necessary as an overall benefit and for
water management. In tests 1, 2, 5 and 6 the end gun is operated and applies water
outside the crop area for part of the operation as shown in Figure 8. Operating the end
gun on a timer allowing it to turn on only when it is covering cropland would save
additional water and energy.
4) Water pressure regulation for sprinkler and sprays are an important device to insure
uniform pressure along the lateral. Small and large water drop size affects the DU and
12
CU. Smaller water particles evaporate more readily in the wind and larger droplets
can cause the surface soil to seal and crust over. Additional irrigation can then lead to
erosion of the crusted surface.
Test Average
no. depth applied
(in.) (%) (%) Comments
Wind speed greater than the allowable limitation standard.
1 0.71 82 62 (11 mph)
2 0.45 88 80 Below allowable wind speed, < 5 mph
3 0.60 80 62 Below allowable wind speed, < 5 mph
4 0.60 74 69 Below allowable wind speed, < 5 mph
5 0.56 87 82 Ultra sonic flow meter device used/acceptable wind speed.
6 0.53 83 82 Ultra sonic flow meter device used/acceptable wind speed.
7 0.70 79 73 Ultra sonic flow meter device used/acceptable wind speed.
8 0.32 81 74 Ultra sonic flow meter device used. (Wind = 20 mph)
Wind speed greater than the allowable limitation standard.
9 0.55 90 83 Ultra sonic flow meter device used/acceptable wind speed.
10 0.50 89 83 Ultra sonic flow meter device used/acceptable wind speed.
CU and DU Comparison
95
90
85
80
Percent
75
70 CU (%)
65 DU1/4 (%)
60
55
50
1 2 3 4 5 6 7 8 9 10
Test
13
Figure 3. Uniformity Coefficient and Distribution Uniformity for all evaluation tests.
5) Sprinkler types and operating pressure also affect the DU. For example in test 4 and
7, sprinklers are fixed and do not rotate, operated at water pressure of 25 psi, the CUs
are 74 and 79% respectively. While for the same soil type, crop and weather
conditions, CUs are 88 and 87% in tests 2 and 5 respectively operated at water
pressure about 35 psi, increasing an average of 11%. Then again, fixed spray streams
produce high instantaneous application rates on a small percent of area. While the
rotating streams produce even distribution over the irrigation area and low
instantaneous application rates.
The high instantaneous application rates with high velocity and large droplets are
detrimental to some soil types causing surface damage by sealing off the soil pore
space at the surface.
6) In test 6, the value of was 83% (less than the acceptable value, 85%) was partially
due to the overhanging canopy of the corn leaves in some of the evaluation zones.
8) From graphs of depth catch vs. distance from pivot in Appendix A, the following was
noted:
In tests 2 and 5, the water depth caught ranged between 0.282 and 0.80 in, and
from 0.267 to 1.50 in respectively. The variations in depth occurred at towers 3, 5
and end gun tower.
In test no. 3, the depth caught in cups ranged between 0.118 and 0.991 in. The
variations in depth occurred at towers 2, 4, 6 and 7.
In tests 4 and 7, the depth caught in cups ranged between 0.222 and 1.082 in, and
from 0.361 to 1.083 in respectively. Variations occurred between towers 3
through 7.
In test no. 6, the depth catch in cups ranged between 0.33 and 0.778 in. The
variations in depth catch took place between towers 3, 6 and end gun tower.
In tests nos. 9 and 10, the depth catch in cups ranged between 0.35 and 0.918 and
between 0.305 and 0.833 respectively.
9) Rotator sprinklers can provide a higher CU than fixed type even in windy condition
as seen in test 1, 7, and 8 where =82%, 79% and 81%, respectively.
Note: Test 8,9 and 10 were conducted after fall harvest with no crop on the field.
10) In tests 1and 8 for the same weather conditions (high wind speed), the Uniformity
Coefficients are almost similar (82 and 81%) respectively. But Distribution
Uniformity are (62 and 74%) respectively. The rotating suspended sprays (test 8)
were more efficient than using sprays mounted on the lateral, this fact is clear
especially in windy condition, where water drop losses are less.
14
11) Systems with rotating drop nozzles had greater Uniformity Coefficient and
Distribution Uniformity than the rotating sprays mounted on the lateral by about 2%
as an average value. Likewise, the drop downs had better uniformity than the fixed
sprays mounted on the lateral by about 14% as an average value.
12) In tests 8, 9, and 10, water pressure from sprays in tower no. 1 are less than other
sprays. This is due to the pressure regulation device default.
(6)
The equation gives an approximate estimation for the CU by using only the value of DU
as a quick calculation. Table (3) shows the comparison between CU calculated from
Heermann and Hein and from the statistical equation. The percentage differences
between the two values are 4% as an average.
Each cup can represent an irrigated area as part of the field; so the volume caught by each
cup is the depth of water times the represented area. The ultra-sonic flow meter device
was used to measure the flow in the system. Total volume of water applied by the system
was compared with the total volume of water caught by the cups to indicate water losses.
15
Table 4 shows the comparison between volume of water measured by ultra-sonic flow
meter and the volume of water caught in the cups.
Results of this comparison are as follows.
1) For field test 5, the water volume measured by ultra-sonic flow meter (Q=619 gpm) is
almost the same volume calculated by catch cans. On the other hand, for the same
evaluation procedure in field test no.2, the water volume recorded from the existing
current meter (Q=750 gpm) is more than the water volume calculated from catch cans
by 14.1%.
2) Also, in field test 7, the water volume measured by flow meter (Q=788 gpm) is more
than the calculated volume by 9.2%. But in the same field test 4, the water volume
recorded from the existing current meter (Q=650 gpm) is more than the calculated
volume by 43.7%.
3) These differences in water volume measured or recorded and calculated are due to:
a. The recorded values from the existing meter device may give more or less
than the actual flow flowing in the system though the Fluxus F601ultra-sonic
flow meter is accurate within 2%.
b. Time required (provided by the farmer or by the Manufacturer’s manual) to
complete one full circle of the center pivot system may be more or less than
the actual time.
c. Water leakages from the lateral pipe in the system as mentioned before in
field tests nos.1, 2, 4, 5 and 7, lost on the ground before reaching the catch
cans.
16
d. The canopy of the crop plant, especially corn, may also affect the volume of
water caught by the cups in the evaluation test if the cups are too close to the
crop line (test no.6).
4) The volume of water applied to the irrigation area can also be found from multiplying
the average depth applied by the irrigated area. The difference between the calculated
volumes from the average depth and the volume calculated from each cup caught by
is about 2.7%.
5) The big difference between water volumes applied and volume caught by cups in test
8 comparing with tests nos. 9 and 10 is due to the high wind speed.
Where:
Table 5, shows the values of for the evaluation tests where the ultra-sonic flow
meter is used.
Test PELQ
No. (%) (%)
5 82 86
6 82 75
7 73 68
8 74 50
9 83 85
10 83 73
The values of PELQ for tests 5, 6, 9 and 10 are acceptable. The low value of PELQ
is usually associated with poor system design and management problems (except for the
condition of operating in windy-test no.8). Meaningful comparisons between several
17
systems modifications or methods can be made by comparing values of PELQ. For
a proper system operation, PELQ should be similar to the value of DU.
Calculations
18
was then multiplied by the pounds of N that would have been applied if the system
operated at 100% uniformity.
In order to rate the distribution uniformity of nitrogen, the average deviation was
calculated. At every catch can, the depth of water was translated to an applied rate of
nitrogen in lbs/acre. This was then compared to the average application rate (65 or 200
lbs/ac) and then divided by the total number of catch cans giving the deviation of the the
whole data set from the average intended application. Results follow.
400
350
Nitrogen (lbs/acre)
300
250
200
150
100
50
0
0 200 400 600 800 1000 1200 1400
Distance from Center Pivot (ft)
Figure 4. Nitrogen distribution based on water applied and sprinkler system uniformity in test 4.
19
Test 4, 65 lbs Nitrogen
140
120
100
Nitrogen (lbs/acre)
80
60
40
20
0
0 200 400 600 800 1000 1200 1400
Distance from Pivot Center (ft)
Figure 5. Nitrogen distribution in test 4 assuming 65 lbs/acre applied over the season by fertigation in
test 4.
200
Nitrogen Applied (lbs)
150
100
50
0
0 200 400 600 800 1000 1200 1400
Figure 6. Graph based on 200 lbs nitrogen application over a season via fertigation in test 4.
20
Table 6. Number of acres that receive different amounts of Nitrogen in Test 4.
Figure 6 shows how the center pivot applies nitrogen (blue) based on water collected
from the cans. The red line indicates the amount of water applied if uniformity were at
100%.
350
300
Nitrogen (lbs/acre)
250
200
150
100
50
0
0 200 400 600 800 1000 1200 1400 1600
Distance from Pivot Center (ft)
Figure 7. Nitrogen application was calculated on test 2 with the greatest uniformity rating (89%)
assuming all nitrogen is applied via fertigation (200 lbs. over the season).
In tests 2 and 4, each point was compared to the average applied lbs/acre of
nitrogen to attain the deviation.
Test 2 – CU (89%), Nitrogen Deviation (10%)
Test 4 – CU (73%), Nitrogen Deviation (20%)
Overall, there is a 16% difference in uniformity coefficient and a 10% difference between
nitrogen applications. This indicates the negative effects low uniformity has on field
crops. Figure 5 illustrates that some areas of the field will receive upwards of 130
lbs/acre while areas near the end gun (which also cover more acres) will only receive 25
21
lbs/acre. Add in a sloping topography and actual nitrogen distribution will decrease
significantly. Thus, it is crucial to regularly maintain a system and perform a type of
uniformity test or check to assure appropriate applications.
Similar to the water distribution graphs, there are certain areas where leaks
(Figure 7 at 420 and 1,000 ft) or irregular watering patterns occur. The higher
application rate in these sections could leach into the ground water and be lost to the plant
or collect in runoff. Runoff, which would take nitrogen away from the higher elevations,
would then pond and could percolate out past the plant roots making it unavailable.
Uniformity therefore has the potential to cost the farmer several times over when
fertigation is used in a system with poor uniformity.
Recommendations
Distribution Evaluations
1) Evaluation test should always be done in open area and catch cans should be far from
the canopy of the crop as shown in Figures 4 and 6.
2) Following ASAE S436.1standard it is suggested to run systems when winds are low,
preferably less than 5 mph and no greater than 11 mph.
3) Regular system maintenance is necessary including repair, adjustment or modification
to keep the system operated efficiently. If CUs are periodically measured (at least
annually) system repairs and adjustments can be scheduled when coefficients fall
below the desired values. This will save operation costs and conserve water.
4) Water pressure should be tested at the sprinkler outlet to ensure that each sprinkler
operates at the design pressure especially sprinkler(s) which give low or high volume
caught in catch cups, which affects the overall DU and CU.
5) Low uniformities in the center pivot system have compounding negative effects when
nitrogen is applied. Therefore, any major leaks and poor end gun performance need
to be fixed or adjusted to insure the highest uniformity possible.
22
12) Operating of the pressure regulation for all sprays should be checked and replaced
when needed.
Acknowledgements
The authors want to thank Lyndon Kelley for his generous contribution.
This work is supported by Biosystems and Agricultural Engineering Department.
The writers thank Mark Sackrider, Ed Groholske , Ryan Groholske and Sally and
Dale Stuby for their coorporation performing evaluation tests in their field.
The writer thanks the staff in the research shop /Department of Biosystems and
Agricultural Engineering for their help with supporting the team work with tools.
Finally the writer thanks Daniel Morgan for his support doing fields tests and Sean
Woznicki for preparing the maps.
23
References
1- ANSI/ASAE S436.1, “Test Procedure for Determining the Uniformity of Water
Distribution of Center Pivot and Lateral Move Irrigation Machines Equipped with
Spray or Sprinkler Nozzles”, Jun1996 (R2007).
2- ASABE, Resouce, Engineering and Technology for a Sustainable World, Vol.18,
No. 5, September/October, 2011.
3- GW Asough and GA Kiker “The Effect of Irrigation Uniformmity on Irrigation
water Requirements” Water SA Vol. 28, No.2, April 2002.
4- Jack Keller and Bleisner RD “Sprinkler and Trickle Irrigation”, Van Nostrand
Reinhold, New York, 1990.
5- MAD Estimated Water Withdrawals for Irrigated Crops
in Michigan, by Crop Type, 2006.
6- Merriam, John L., and Jack Keller, “Irrigation System Evaluation and
Improvement” Utah State University, Logan , Utah, 1973.
7- Michael D. Dukes and Calvin Perry “Uniformity testing of variable-rate cenetr
pivot irrigation control systems” Precision Agric. Vol.7, pp205-218, 2006.
8- MSU Excell spreadsheet for the evaluation test, prepare by London Kelley.
9- Nelson irrigation Corporation “ water application solutions for Ccenter pivot
irrigation” 2004.
10- 10- Jerry Wright and et al “Nitrogen Application with irrigation water-
Chemigation”, University of Minnesota/Extension, clean water series, 2002.
24
Appendix A
Figure 9. Catch cups lined up along the edge of the corn crop.
25
Figure 10. Measuring water volume from catch cup.
Figure 11. Row of cups before irrigation line has passed over.
26
Figure 12. Using the Ultra-sonic flow meter to measure flow rate through the
system.
27
Figure 14. Water leaking from the lateral joint on the center pivot line.
28
Figure 16. Rotating sprays suspended from the lateral of the center pivot.
29
Appendix B
30
Figure 20. Water distribution in test 3.
31
1.6
Tower 2 Tower 3 Tower 4 Tower 5 Tower 6 Tower 7 End Gun
1.4
1.2
Container Catch Depth (in)
1.0
0.8
0.56
Ave. in
applied
0.6
0.4
0.2
0 500 1000 1500
Test no.5
Figure 22. Water distribution in test 5.
0.9
0.7
Container Catch Depth (in)
0.6
0.53 in.
0.5 Ave.
applied
0.4
0.3
0 200 400 600 800 1000 1200 1400
Distance from Pivot (ft)
Test no.6
Figure 23. Water distribution in test 6.
32
1.2
1.1 Tower Tower Tower Tower Tower 6 Tower 7 Tower End Gun
2 4 5 8
1.0
0.9
Container Catch Depth (in)
0.8
0.5
0.4
0.3
0 500 1000 1500
Distance from Pivot (ft)
Test no.7
0.6
0.4
Container Catch Depth (in)
0.3
0.32 in.
Ave.
applied
0.2
0.1
0.0
0 200 400 600 800 1000
Distance from Pivot (ft)
Test no.8
33
1.0
0.8
Container Catch Depth (in)
0.7
0.6
0.55 in.
Ave.
0.5 applied
0.4
0.3
0 200 400 600 800 1000
Distance from Pivot (ft)
Test no. 9
Figure 26. Water distribution in test 9.
0.9
Tower 1 Tower 2 Tower 3 Tower 4 End Gun
0.8
0.7
0.6
Container Catch Depth (in)
0.3
0.2
0.1
0.0
0 200 400 600 800 1000
Distance from Pivot (ft)
Test no. 10
34
Nitrogen Distribution in Center Pivot
120
100
Nitrogen (lbs/acre)
80
60
40
20
0
0 200 400 600 800 1000 1200 1400 1600
Distance from pivot center (ft)
Figure 28. Distribution of nitrogen when applying 65 lbs/acre nitrogen fertilizer by irrigation in test
2.
35
Appendix C
Table 7. Soil type analysis for each evaluation field
36