3.1 Identification of Trial Farms
3.2.1 Water Metering
3.2.2 Soil Sampling
3.2.3 Calculation of Indicators
3 Methodology
3.1 IDENTIFICATION OF TRIAL FARMS
Six farms were selected to field test the indicators of sustainable irrigation. The farms were divided into three pairs each representing a different combination of enterprise type and irrigation system. All farms were located within the Central Canterbury Plains primarily for ease of access, although it has been estimated that this region contains the second largest area of irrigated land (20% of total) after Mid Canterbury (30% of total)2. Farms were selected from the arable and dairy sectors as these represent the major irrigation users, approximately 30% and 20% of irrigated land area respectively. Furthermore, there was a perception building in the area that irrigation of dairy farms may not be sustainable due to environmental problems such as nutrient leaching.
The three pairs of farms were as follows:
Dairy farms with border strip irrigation, water drawn from north side of Rakaia
1A An established farm with conventional narrow borders (12 m wide)
1B A new conversion with wide borders (24 m wide)
Dairy farms with spray irrigation, water drawn from groundwater in Dunsandel area
2A A recent conversion with travelling irrigators (Briggs Rotorainers & Traymark)
2B A new conversion with centre pivot irrigation
Arable farms with spray irrigation in Kirwee - Greendale area
3A A long-term irrigated farm with siderolls and handshift pipes.
3B A newer system with travelling irrigators (Briggs Rotorainer) and a sideroll
These six farms were chosen to provide a fair representation of the likely diversity of irrigation system types and management practices. It was also hoped that this would give a good spread in the quantitative values of the indicators and the likely problems to be encountered in their determination.
One farm in each pair (1A, 2A, 3A) was chosen because it was already involved in another sustainability project and it either represented ‘standard’ practice or had a well established history of irrigation. Two of the farms (2A and 3A) have been part of a sustainability trial3 operating in the Selwyn District since February 1995. That project developed and tested a broad range of indicators related to bio-diversity, water quality, soil quality, nutrient and energy inputs, and economic and physical outputs. The two dairy farms (1A and 2A) were also involved in the trials for the energy indicators project4. These three farms effectively acted as controls.
The second farm in each pair was chosen by finding a similar farm in terms of location, enterprise, and water source, with an irrigation system that would be amenable to implementation of the Best Management Guidelines. In the case of the two dairy farm groups this also represented ‘newer’ and supposedly more efficient technology, i.e. the use of wide borders and the centre pivot system.
3.2 COLLECTION OF DATA
Each farm was visited in early June 1998 and the necessary data collected by interviewing the farmers. For the dairy farms the following information was collected for the farm as a whole:
- total farm area (ha);
- effective farm area (ha);
- total irrigated area (ha);
- irrigation system type
- milk solids production (kg Milk Solids);
- annual net operating profit after tax ($);
- total quantity of water applied (m3) or sufficient information regarding application depth, area and number of irrigations for this to be estimated;
- annual energy use (where applicable) (kWh);
- labour requirements for irrigation;
- maximum daily water abstraction rate (m3/day);
- fertiliser use [N, P, K, S] (kg/ha);
- agri-chemical use (l /ha or kg/ha).
For the two control dairy farms (1A and 2A) most of this information was available for the 1996/97 and 1997/98 seasons or could be projected. For the two first year conversions (1B and 2B) this information was only available for the 1997/98 season. No financial information was available for the current season.
For the two arable farms similar data was collected but rather than considering the whole farm five paddocks were selected and analysed for the 1998 and 1997 seasons. Due to the lack of meaningful financial information from the dairy farms it was decided that the arable farm surveys would concentrate more on this aspect. Gross margins were determined for each crop as the financial indicator as it was not possible to perform any meaningful comparisons at a whole farm level.
3.2.1 Water Metering
It had been the original intention of the project to install water-metering devices on all the properties. However, this proved to be very difficult for a number of reasons.
On the surface irrigated properties metering was to be accomplished by using water level recorders in the turnout basins and calculating the flow based on the measured opening of the sluice gates.
On property 1A this was complicated by the fact that the turnout design was not straightforward. About 60 ha of the border strip area was watered directly from one turnout basin. This basin also fed two other main races. One flowed down to another border strip irrigated area of the farm of approximately 50 ha and then on to another farm. The diesel pump for the big gun irrigator also drew from this race. The other main race branch also flowed on to another farm in the scheme. About 15 ha of property 1A was irrigated from this race through a secondary turnout.
On property 1B flow metering was installed by Lincoln Environmental in mid February meaning that only two irrigation cycles were monitored. The monitoring showed that the flow rate on to the property varied considerably during each irrigation. It is believed that this was most likely a result of the build up of debris at the inlet structure (McIndoe, pers. com).
On the spray irrigated properties none of the head works were set up with a sufficiently straight length of pipe between the wellhead and the first branch. The usual recommendation for flow meters to perform accurately without being unduly influenced by the effects of turbulence is that there be a straight length of pipe of ten diameters upstream and five diameters downstream. Property 2A was complicated by having two wells feeding into the same multiple ring-main system. Having some water pumped into the system from a neighbouring property complicated property 2B. On this property Lincoln Environmental measured the flow by measuring the pressure at the centre pivot. A short-term experiment was performed to calibrate pressure against flow using an ultrasonic flow meter borrowed from the Canterbury Regional Council. While properties 3A and 3B each had only one pump, metering was complicated by the main line branching at less than 15 diameters from the wellhead. In each case more than one meter would have been required for accurate measurements.
Due to the difficulty of obtaining accurate in-situ flow measurements it was decided for the purposes of this study to determine total water application volume based on the irrigated area, the typical depth of application, and the number of irrigation cycles. It is acknowledged that this is not the most accurate method of determining total water input. However, it is believed that equally inaccurate results would have resulted from installing single flow meters under these sub-optimal conditions, and the extra cost of installing extra flow meters on every property could not be justified.
3.2.2 Soil Sampling
On each farm soil sampling was carried out in accordance with the protocol established with Crop & Food CRI for Soil Aggregate Stability Tests. Using a 75 mm diameter coring tool, five samples were taken on a transect from each sampled paddock, to a depth of 75 mm. The five samples were then bulked together into one bag to provide enough soil to carry out tests for Aggregate Stability Tests, pH, total organic carbon and nitrogen. On the dairy farms five paddocks were sampled to give a representative cross-section of the farm, except for farm 2B where only four paddocks were sampled. On the arable farms samples were taken from each of the five paddocks for which gross margins were calculated.
All soil tests were carried out at the Crop & Food CRI soils laboratory at Lincoln using standard techniques.
Aggregate stability: Samples of 2-4 mm air-dried aggregates (50g) were shaken over a stack of sieves using a wet-sieving apparatus. Sieves contained 2.0, 1.0 and 0.5 mm diameter openings. Stable aggregates remaining on each sieve were weighed after oven drying at 105°C, and aggregate stability was expressed as mean weighted diameter (MWD) in millimetres.5
pH: Samples of oven-dried soil (12 g) were shaken with 25 ml of distilled water. After settling overnight, pH measurements were made using a HANNA pH meter with a combination electrode.6
Total organic carbon and total nitrogen: Determinations were made on individual 0.4 g oven-dried soil samples by an automated combustion (Dumas) method using a LECO CNS-2000 analyser operating at 1050°C.7
Water Holding Capacity (Readily Available Water): The water holding capacity was determined by comparison of neutron probe and quick draw tensiometer readings using the methodology developed by MAF Technology.8
3.2.3 Calculation of Indicators
Calculation of most indicators followed from the input data in a straightforward manner. On farm 1A, which had surface and spray irrigation, the energy use and labour indicators were calculated for each system separately and for the property as a whole.
On farm 2B with the centre pivot system the labour requirements were determined separately for two parts of the season. For the first 90 days the system completed one revolution of each paddock in 24 hours (applying 25 mm) and was then shifted (every day) to the next paddock on a six-day cycle. For the last 120 days of the season the system completed two revolutions of each paddock in 48 hours (applying 50 mm) and was then shifted (every second day) on a twelve-day rotation.
This project has concentrated on those indicators that can be determined from historical data. Some information on daily indicators has been collected in the companion project run by Lincoln Environmental on farms 1B and 2B. As an alternative to the recommended indicator of daily percentage of water stored in the root zone, it was decided to try and estimate the ratio of the seasonal water application to the seasonal water demand as seasonal reference evapotranspiration with and without rainfall correction. The ratio of total water depth applied (D) to potential evapotranspiration demand (ET) and rainfall corrected water demand (ET*) were calculated.
The rationale for the selection of these ratios is as follows. The intent of the original indicator was to determine the efficiency of water application in terms of that which remains within the root zone for later extraction by plants. Water moving below the root zone will be unavailable and is effectively wasted. The only way to accurately monitor this is to use real-time soil moisture monitoring to construct a mass balance for various layers within the soil profile. In an ideal irrigation system no irrigation water would move below the root zone meaning that the amount of water applied would exactly match the evapotranspiration demand less any effective rainfall. Effective rainfall in this context is that which itself remains within the root zone. Ideally all rainfall is effective, however when the total amount of rainfall exceeds the current water storage capacity of the soil (difference between current water content and field capacity) some must either drain or run off.
In lieu of costly real-time monitoring it was believed that these ratios would provide at least some indication of how well the design and management of the irrigation systems was matched to actual water demand. The hydrological design of irrigation systems is by no means an exact science9, however most design recommendations are based on satisfying likely average potential evapotranspiration demand during the summer period in the absence of significant rainfall. Thus, the ratio D/ET was expected to reveal useful information about how closely the actual seasonal management of the irrigation systems conformed to the potential evapotranspiration demand. Ratios of around one would imply that the systems were being operated close to the design situation. The ratio D/ET* then relates to the total amount of water applied compared to the water demand adjusted for rainfall. A ratio close to one would imply that the system was being managed to use effectively all the available rainfall. This will be extremely difficult to achieve in practice, as it requires that the irrigation system is infinitely flexible in operation, to stop and start at will at any point in the system, and that all rainfall is effective. That is, it requires the farmer to anticipate each rainfall event and that the soil will be dry enough to accommodate all rainfall within the root zone.
Seasonal evaporation and rainfall data from Lincoln University were used in this study. Evapotranspiration data quoted is reference potential evapotranspiration calculated using the Penman formula. For the 1996/97 season the total evapotranspiration (ET) from the beginning of October to the end of April was 770 mm and the corresponding rainfall (P) was 345 mm, making the effective water demand (ET* = ET - P) 425 mm. For the 1997/98 season the corresponding figures were ET= 935, P= 160, ET*= 775. Note that these figures are to the nearest 5 mm. This analysis was not performed for the arable farms because it was too difficult to evaluate accurate or meaningful values of ET that corresponded with the actual crop water requirements.
3.3 IRRIGATION SYSTEM AUDITS
For the two control farms with spray irrigation (2A and 3A) an irrigation system audit was carried out as part of this project. The purpose of the audit was to determine the technical efficacy of each system and the degree to which its overall performance might be increased. This related to the design and maintenance of the irrigation system as discussed in section 2.4. This information may be useful in explaining some of the indicator values. Audits on the other two spray irrigated farms (2B, and 3B) were carried out by Lincoln Environmental as part of the Best Management Guidelines project.
Details of the irrigation systems of the two farms were digitised onto a computer. Each component from the system, including pipe size and layout, pumps, water source characteristics, valves and outlet pressure and flow relationships, were entered using the IRRICAD™ software. Once the auditor was satisfied that IRRICAD™ was correctly predicting what was happening in reality, the irrigation system was computer analysed. Analysis involved running a series of different irrigation and management options and recording the effect that these had on the pressure and flow characteristics of the outlets and pipes, plus pump and well performance.
2 Bright, J.C. & Wells, C.M., 1997. A Survey of Farmers’ Approaches to and Perceptions about Irrigation Management. Report to MAF Policy. Report No 2773/1, Lincoln Environmental. 3 Selwyn Stewardship Monitoring Project. Selwyn District Sustainable Agriculture Society (Inc.) 4 Wells, C.M., 1998. Total Energy Indicators of Agricultural Sustainability. Report to MAF Policy, Agriculture New Zealand Ltd, Christchurch. 5 Kempere, W.D. & Roseneau, R.C., 1998. Aggregate stability and size distribution, In: Klute, A., ed. Methods of soil analysis, Part 1. Physical and mineralogical methods. 2nd edition. Soil Science Society of America, Madison, WI. pp425-442. 6 Blakemore, L.C., Searle, P.L. & Daly B.K., 1987. Methods of chemical analysis of soils. New Zealand Soil Bureau Scientific Report 80. 7 McGill, W.B. & Figueirdo, C.T., 1993. Total nitrogen. In: Carter, M.R. ed. Soil sampling and methods of analysis. Lewis Publishers, Boca Raton, FL. pp 659-662. 8 McChesney, I.G., 1990. Irrigation Scheduling Notes. Irrigation Scheduling Service, MAF Technology, Lincoln. 9 The primary author taught hydrological design for a number of years at Massey University.
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
