5.1 Irrigation System Description
5.2.1 Soil Moisture Measuring Equipment
5.2.2 Flow Measuring Equipment
5.2.3 Irrigation Management Spreadsheet
5.4 Data Collection and Analysis
5.6.1 Milk Production
5.6.2 Seaonal Water Use
5.6.3 Water Stored In The Root Zone
5 Border Strip Irrigated Dairy Farm
5.1 IRRIGATION SYSTEM DESCRIPTION
The total area of the border-strip farm is 200 ha, with an effective farm area of approximately 192 ha. Of this, approximately 165 ha is
The irrigation system is a border strip system with 24 metre wide strips. These strips are twice the width of the traditional border strip design. Length of borders range from about 200 metres to 300 metres.
The water for the system is diverted from the Rakaia River, approximately three kilometres from the farm boundary. The supply enters the property at the western end on Knyvetts Road. A network of headraces supplies water to the paddocks. Generally, three or four strips are watered at the same time.
The layout of the system is shown in Figure 3.
Figure 3. Layout Of Border Strip Irrigation System
5.2EQUIPMENT INSTALLATION
5.2.1 Soil Moisture Measuring Equipment
In January 1998, an Aquaflex™ unit with four sensing cables had been installed in Paddocks K and L (2 sensors in each paddock connected to a central logger). The computer chip in the data logger on this unit was upgraded to allow data to be read and downloaded by Andrew Mulholland using a Palm Pilot hand-held computer. This gave Andrew the ability to take readings at any time.
Because this system had provided excellent results in the 1997/98 irrigation season, further soil moisture equipment was not considered necessary.
5.2.2 Flow Measuring Equipment
In October 1998, the water level recorder used in the 1997/98 irrigation season was re-installed just upstream of the permanent gate structure on the main supply channel at the entrance to Andrew Mulholland’s farm. This was the same location as had been used in the 1997/98 season.
At this point, flow was delivered to the main channel through a main gate, or by closing the main gate through a side channel used to water a small area of the property at the corner of Knyvetts Road and Main Rakaia Road (Paddock T).
As for the previous season, the basic equation used for the calculations was:
Flow = 0.41 x (Ws-0.2 x H) x g 0.5 x H 1.5 where
Ws = width of gate
H = height of water upstream of gate measured from main sill height
g = acceleration due to gravity
For flow through the side gate, H was adjusted to H - Hs, where
Hs = height of side sill above main sill.
Lincoln Environmental downloaded the data from the level recorder.
5.2.3 Irrigation Management Spreadsheet
An irrigation management spreadsheet that carried out a daily water budget based on rainfall, irrigation water applied, and average Lincoln daily evapotranspiration and similar to the one provided to Bruce Rolls was supplied to Andrew Mulholland. Again, the use of average Lincoln daily evapotranspiration allowed a meaningful forward projection of soil moisture deficits to be made.
The spreadsheet had facilities for Andrew Mulholland to enter actual daily evapotranspiration (published weekly in the Christchurch Press), and actual rainfall (measured by Andrew on the property). It also had the ability to convert Aquaflex™ soil moisture readings into soil moisture deficits and update the soil moisture status of the spreadsheet.
Running totals of soil moisture deficits based on soil moisture readings and on daily water budget were kept in the spreadsheet to allow a comparison to be made between the two methods.
5.3 IRRIGATION PLAN
Andrew’s Mulholland’s irrigation system differs from groundwater supplied systems in that the water supply from the Rakaia River is subject to water restrictions that change at the beginning of each month.
Water for the system is diverted from the Rakaia River through a main canal. The supply is subject to the rules for abstraction from the Rakaia River under the terms of the Rakaia National Conservation Order (NCO). Water cannot be diverted when flow rates in the Rakaia River are lower than the one in five year monthly low flows, with sharing rules applied above the minimum flow levels. There is no real way of knowing when restrictions might occur, except that the likelihood of restrictions increases as the irrigation season progresses.
The uncertainty involved in water supply meant that Andrew was forced to plan to keep soil moistures as high as possible, particularly at the end of each month.
Based on information from the previous season, Aquaflex™ readings in Paddock L were to be used for scheduling purposes. This was because the water holding capacity of the soil in Paddock L was lower than in Paddock K, and represented the soil moisture status of the majority of the property. An Aquaflex™ reading of 14.0 was used as the minimum allowable before irrigation took place. This reading was established by visual observation of pasture growth on the more stony areas of the property and Aquaflex™ readings when water stress occurred on these areas.
Because the border-strip irrigation always returned soil moistures to field capacity with very little control over the depth applied, the timing of irrigation was more important to the decision-making process than the depth applied.
Andrew had carried out some pre-season work to ensure that the system operated as efficiently as possible. The work included gate maintenance, sill adjustment, race cleaning, and improvements to the intake at the Rakaia River. Incorrect sill levels are one of the major problems, largely because they were not well constructed initially.
5.4 DATA COLLECTION AND ANALYSIS
5.4.1 Soil Moisture Data
The collection of soil moisture data, interpretation of the data and management of the irrigation system were carried out by Andrew Mulholland, with very little input from Lincoln Environmental.
Andrew Mulholland used the Palm Pilot to either read or download the soil moisture information on a regular basis, and graphed soil moistures and entered readings into the spreadsheet to monitor and plan the irrigation. As with the spray irrigated dairy farm, this generally worked very well, except that the battery life of the Palm Pilot was too short. Reloading the software onto the Palm Pilot was carried out by Lincoln Environmental staff.
5.4.2 Flow Measurement Data
The water use data, which was calculated from the water level recorder data, was collected and interpreted by Lincoln Environmental. Andrew Mulholland assumed that all irrigation applications applied 90 mm of water (based on measurements taken in 1997/98), and entered this value into the spreadsheet.
The water level recorder stopped working over December and January (it was flooded by very high water levels), and water use estimates based on Andrew’s spreadsheet entries of irrigation operating times were made to cover this period.
5.5 IRRIGATION MANAGEMENT
5.5.1 Operation
Andrew Mulholland started irrigation on 22 September 1998, and irrigated again ten days later on the basis of soil moisture measurements. Because the farm can be irrigated within a three day watering cycle, there was a need to project ahead only three days, simplifying the operation.
Rainfall in October allowed him to delay the next irrigation until 23 October. He then irrigated on a regular basis through to mid January, initially on a 17 day cycle but reducing that to 8 days throughout December and the first half of January.
Rainfall in January and February and March allowed the return interval to be extended to about 15 days. No irrigation was applied in April, (the last watering at the end of March carried soil moistures through) but the dry April/May period resulted in another irrigation in mid May and another at the beginning of June. This last irrigation was stopped before completion because of rainfall.
Irrigation into May and June is extremely rare, as normally the last irrigation for the season occurs in March or April.
Although flow was available for most of the 1998/99 season, there were four periods when water, under the terms of the NCO, could not be diverted. On one occasion, in February 1999, this caused soil moistures to fall well below the point where irrigation would have taken place had water been available. Pasture stress was evident on the lighter soils, although the heavier soils tended to hold on quite well.
5.5.2 Maintenance
The sill maintenance carried out during the winter was lost in the first few irrigations, mainly because the grass hadn’t stabilised as well as expected. A day was spent on sill maintenance during the season.
Three dams are leaking, and will require maintenance before next season.
Other than the problems mentioned above, the system performed very well.
5.6 END OF SEASON INDICATORS
Three key indicators have been calculated. They are:
- cubic metres of water used per kilogram of milk solids produced (m3/kgMS);
- percentage of water flowing onto the farm (irrigation and rainfall) that is stored in the root zone (%);
- depth applied / (evapotranspiration – rainfall), which is a measure of water applied versus water required.
The information required to calculate the indicators is given below.
5.6.1 Milk Production
Andrew Mulholland’s milk production for the season was 157,000 kg MS.
5.6.2 Seasonal Water Use
Over the irrigation season, 19 irrigation events took place. Depths applied ranged from 85 mm to 145 mm per irrigation with an average of 116 mm. Volume of water used was as follows.
Pre December (measured) 646,100 m3
December – January (estimated) 1,649,300 m3
Post January (measured) 1,451,100 m3
Total water used 3,746,500 m3
5.6.3 Water Stored in the Root Zone
By examining the soil moisture deficit immediately before irrigation, it was possible to calculate the amount of water stored in the soil and compare it to the amount of water applied for each irrigation or rainfall event. With the border strip irrigation, it was assumed that the soil was always returned to field capacity by the irrigation because the applied depth was always considerably larger than the soil moisture deficit. It was not necessary to calculate soil moisture deficit immediately after irrigation.
The equation for calculation of soil moisture deficit used in the 1997/98 study was recalibrated, because the comparison with the water budget showed that soil moisture deficits as calculated by the previous equation were larger than was realistically possible. The recalibrated equation was:
Soil moisture deficit = 3.5 x Rt + 38 x Rb –645 mm
where Rt and Rb are the top and bottom readings for the Aquaflex™ sensor readings respectively.
The soil moisture deficit at the stress point was again calculated to be about 60 mm and the total water holding capacity probably greater than 100 mm.
A summary of the data for Paddock L is as follows:
Total irrigation water applied 2278 mm
Total irrigation drainage 1500 mm
Irrigation water stored in the root zone 778 mm
Total rainfall 358 mm
Total rainfall drainage 59 mm
Rainfall water stored in the root zone 299 mm
Total evapotranspiration (Lincoln) 973 mm
It is interesting to note that the total water stored in the root zone differed from the seasonal evapotranspiration by only about 100 mm. This difference could easily be explained by differences in evapotranspiration between Lincoln and the property, and errors in flow measurement and soil moisture measurement.
5.7 SEASONAL INDICATORS
Cubic metres of water used per kilogram of milk solids produced 23.9 m3/kgMS
Percentage of irrigation water stored in the root zone 34 %
Percentage of rainfall stored in the root zone 83 %
Depth applied / (Evapotranspiration – Rainfall) 3.7
More water was used than last year, primarily because improvements o the intake at the Rakaia River were made, allowing a higher, more uniform flow to be taken.
Milk production was also higher, so the cubic metres of water used per kilogram of milk solids produced overall was similar to last year.
The percentage of water stored in the root zone has fallen from 45 percent last year to 34 percent this year. This can be attributed partly to the higher flow rates and depths of water applied and partly to the need to irrigate at sub-optimal soil moistures because of the high probability of water restrictions in the days following.
There is a trade-off between potential loss of production and water use efficiency. In Andrew Mulholland’s case, the financial cost of irrigating is small and the potential loss of production through not irrigating large. Given the uncertainty of water availability, he had little choice but to irrigate.
Contact for Enquiries
MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND
Fax: +64 4 894 0721
Contact this person
