Appendix 3: The Best Practice Dairy Catchments
The best practice dairy catchments for sustainable growth project commenced in 2001. There are presently four such catchments in those parts of the country where Fonterra operates. These have typically been studied intensively for four to 10 years. They are the Toenepi Catchment near Morrinsville in the Waikato, the Waiokura catchment in South Taranaki, the Waikakahi catchment on the north bank of the lower Waitaki River, and the Bog Burn Catchment in eastern Southland.
These four catchments were selected because they are predominantly dairying, and they cover a range of geographic locations and climate types around New Zealand. Production levels are typically above the national average, most farmers purchase additional feed, and fertiliser use is typical of other dairy farms in New Zealand.
As part of this project a review was carried out of existing monitoring programmes in these catchments, which are summarised below. It is also noted that a joint monitoring programme currently being undertaken as part of the National Catchment Dairy Project (Environment Waikato, NIWA, Dexcel and Ag Research) has undertaken significant work in the Toenepi Catchment which has resulted in a number of Farm Action Plans being developed. This work is based on a significant amount of monitoring information and Fonterra expect that the 'template' will now be utilised within the other catchments. [Another catchment at Inchbonnie near Lake Brunner has recently been added to this programme. As it is outside the Fonterra ambit, however, it is not suitable for including as part of monitoring the benefits of implementing the Accord (unless Westland Dairy comes up with a similar arrangement with their suppliers).]
A wide variety of research work has been undertaken in the four best practice dairy catchments. It is very important to note that this work has focused on monitoring of water quality, and best practice dairying through intensive studies linking environmental outcomes, farm inputs and farm productivity and economics. In no case, however, has this work directly focused on the benefits of implementing the Accord. In the following reviews of the work undertaken to date we only discuss the work with an environmental emphasis.
1 The Toenepi Catchment
The catchment
The Tonepi catchment covers some 15.5 km2 of rolling country towards the east of the Waikato, near Morrinsville. Almost two thirds of the catchment is in dairying with 21 farms present. The balance of the catchment is more or less equally divided between dairy run-off and drystock. In 2001, the average herd size was 211 and the stocking rate was three cows/ha, and no stock were wintered off-site.
Average rainfall is 1200mm, mean flow is 70l/s, and there is 39.4 km of stream length present. Of this, 56 percent is permanently fenced on one side, six percent permanently fenced on both sides with 25 percent of stream length both fenced and the riparian zone planted. Eighty-six percent of the farmers in the catchment dispose of their dairy effluent to water after pond treatment, with the balance irrigating to land.
Monitoring to date
The Toenepi catchment has been the longest studied of the four best practice catchments, having been studied since 1995. Weekly sampling of water quality at three sites in the catchment was carried out for two years from March 1995 to April 1997. Sampling frequency was reduced after this, but monthly monitoring of many variables continues to this day. A permanent flow recording site is present at the lower end of the catchment.
As in all four of the best practice catchments, there is a vast amount of water quality information available from analyses of all the water quality samples collected in the catchment. This information is summarised in Wilcock and Duncan (2003). This work, which has taken place over the last 10 years, generally shows water quality to be significantly degraded, more so than in the other three catchments.
In broad terms, although levels of suspended solids are generally low, they are sometimes elevated as is turbidity. Nutrients are elevated, and in the case of phosphate, increase downstream. Specific yields of nutrients are the highest of the four catchments, which may well reflect the higher dairy stocking rates in the Toenepi versus the other catchments. Specific yields of suspended sediment and nutrients dropped, typically between 20-40 percent, between the years 1995-97 versus 2001/02. Levels of E. coli are high, and regularly exceed contact recreation guidelines. There are occasional records of low dissolved oxygen and high levels of ammonia nitrogen.
In spring 2001, as in the other best practice catchments, soil physical, chemical and biological quality assessments were made of the major soil types within the catchment (Mongahan et al 2001). About two-thirds of the soils in the Toenepi were compacted, and Olsen P, which averaged 53, was considerably higher than the target of 25-35.
This work was repeated in winter 2003 (Drewery et al 2003). At this time, soil compaction was evident on 43 percent of farms, and again Olsen P was high on most properties, exceeding 40 in 70 percent of the properties in the catchment. Many sites had lower than optimum soil potassium.
There are three main sources of phosphate entering the water courses in the catchment, with each of soil loss, excessive phosphate fertility and discharges from dairy effluent ponds each making up approximately 30 percent of the phosphate running off farms.
2 The Waiokura Catchment
The catchment
The Waiokura catchment rises on the south Taranaki ring plain near Kaponga and flows south to enter the sea near Manaia. Mean flow is 313 l/s. In 2001, there were 17 dairy farms in the catchment with an average herd size of 256 and an average stocking rate of 3.4 cows/ha. Twenty-nine percent of stock are wintered off-site.
Two-thirds of the farmers in the catchment irrigate to land, with the other third discharging to water after pond treatment. Of the total stream length, 38 percent has riparian fencing on both sides, 31 percent on one side and 31 percent of the streams are both fenced and planted out with riparian species.
Monitoring to date
Intensive work has been carried out in the catchment since 2001. Fortnightly sampling of water quality is carried out at three sites in the catchment by the TRC, and a permanent flow recording site is present at the lower end of the catchment.
As in all four of the best practice catchments, there is a vast amount of water quality information available from analyses of all the water quality samples collected in the catchment. This information is summarised in Wilcock and Duncan (2003). This work, which has taken place over the last five years, generally shows water quality to be significantly degraded.
In broad terms, although levels of suspended solids are generally low, they are sometimes elevated as is turbidity. Nutrients are also somewhat elevated, particularly phosphate at the upper site in the catchment. Levels of E. coli are high, and regularly exceed contact recreation guidelines. Dissolved oxygen is generally high and levels of ammonia nitrogen low.
In spring 2001, as in the other best practice catchments soil physical, chemical and biological quality assessments were made of the major soil types within the catchment. Olsen P, which averaged 65, was much higher than the target of 25-35. The volcanic soils in the catchment are not subject to compaction.
This work was repeated in winter 2003. Olsen P was high on almost all properties, exceeding 40 in 95 percent of the properties in the catchment. Many sites had lower than optimum soil potassium.
There are three main sources of phosphate entering the water courses in the catchment. Excessive phosphate fertility makes up about 40 percent of this, whereas discharges from dairy effluent ponds and soil loss each making up nearly 30 percent of the phosphate running off farms.
3 The Waikakahi Catchment
The catchment
The Waikakahi catchment covers predominantly flat pasture on the north bank of the Waitaki River, which it enters near Glenavy on State Highway 1. The stream is sourced predominantly by several large springheads, along with some run-off from the adjacent hill country. Mean flow is 537 l/s, although summer mean flow is 1,850 l/s.
The catchment is atypical in that it is part of the Morven-Glenavy-Ikawai irrigation scheme, which takes water from the Waitaki River. Of the 4,100 ha of flat land in the catchment, about 95 percent is irrigated, 80 percent by border-dyke. This predominantly border-dyke irrigation means that stream flows are much higher in summer (typically four to nine times winter flows) when irrigation by-wash leads to substantial overland run-off to the stream. Catchment soils are light silt loams - they are gravel outwash soils from the former bed of the Waitaki - but have been greatly improved with irrigation and the use of fertiliser. While average catchment rainfall is only about 540 mm per annum, another 810 mm is applied on average via irrigation.
Dairying 'took off' in the catchment in the mid 1990s. While there have long been several dairy farms present, there are now 11 dairy farms which cover almost all the lowland parts of the catchment. The average dairy farm size in 2001 was 244 ha with an average herd size of 665 and a stocking rate of 2.8 cows per hectare. Almost two-thirds of these stock are wintered off-site. Three-quarters of the farms dispose of effluent via pond treatment and irrigation, with the balance equally divided between pond treatment and disposal to water, and direct effluent irrigation to land.
The initial impacts of dairying on stream water quality and stream 'health' were severe. Unimpeded stock access led to significant siltation of the stony-bottomed bed of the stream, and to declining water quality. The value of the stream for trout spawning, and as habitat for invertebrates, juvenile fish and native species, was much reduced.
Since that time major efforts have been undertaken by Environment Canterbury, Fonterra and the farmers to reduce the impacts of dairying, and so restore water quality and stream health. This has included fencing off over 90 percent of the streams and the major springheads, [Although the quality of some of the fencing is not high, being very close to the stream in some instances.] providing stock crossings, and riparian retirement and planting (22 percent of the stream length is both fenced and planted).
Monitoring to date
Intensive work has been carried out in the catchment since 2001, although some earlier work was carried out by the Regional Council Environment Canterbury (ECan). Monthly sampling of water quality is carried out by ECan at one site, and a permanent flow recording site is present at the lower end of the catchment.
As in all four of the best practice catchments, there is a vast amount of water quality information available from analyses of all the water quality samples collected in the catchment. This information is summarised in Wilcock and Duncan (2003).
This work, which has taken place over the last 10 years, generally shows water quality to be significantly degraded. In broad terms, levels of suspended solids are consistently elevated as is turbidity. Nutrients are also elevated, and in the case of phosphate increase downstream, and levels of E. coli are high, regularly exceeding contact recreation guidelines.
In spring 2001, as in the other best practice catchments, soil physical, chemical and biological quality assessments were made of the major soil types within the catchment. Overall soil quality was good, although about a third of the soils in the Waikakahi were compacted, and Olsen P, which averaged 46, was considerably higher than the target of 25-35.
This work was repeated in winter 2003. At this time soil compaction was at a low level, but again Olsen P was high on many properties, exceeding 40 in 65 percent of the properties in the catchment. Many sites had lower than optimum soil potassium.
A modelling study carried out in the Waikakahi and Bog Burn catchments showed that the wintering part of the dairy system made disproportionately large contributions to nitrate leaching from farms. These were 25 percent and 60 percent respectively, despite being only 15 percent of the whole dairy farming area. This was attributed to both nitrogen remaining in the soil from previous activities, and nitrogen being excreted on to forage crops when plant growth is low during winter.
A study was undertaken by Carey et al (2003) of the rate of run-off of irrigation wash water from border-dyke irrigation. The main findings included:
- that although a large amount of irrigation water was applied across the land, the rate that which it passed was too high to allow adequate infiltration
- concentrations of the nutrients N and, particularly, P, and E. coli in irrigation by wash were consistently higher than acceptable critical levels for water quality
- most phosphate was sourced from high soil phosphate levels, with a need to reduce target levels of Olsen P in the soil.
4 The Bog Burn catchment
The catchment
The Bog Burn catchment, which covers an area of 87.6 km2, is in eastern Southland. The catchment rises at an altitude of above 500 m, and is dominated by dairying with some exotic forestry in the headwaters. Catchment soils are heavy and poorly drained. In 2001, there were six dairy farms in the catchment with an average herd size of 748 and a stocking rate of 2.9 cows/ha. All dairy stock are wintered off-site. Forty percent of the stream length in the catchment was fenced on both sides, 40 percent on one side only and 20 percent of the stream length was riparian fenced and planted. All six farms dispose of dairy effluent to land; two do so directly and four do so following pond treatment.
In the year starting January 2002, stream flows varied from 44 to 5186 l/s. Mean flow was 389 l/s and median flow was 163 l/s.
Monitoring to date
Fortnightly sampling of water quality and sediment was carried out by Environment Southland at three sites during 2002. Water quality samples were analysed for suspended sediment, DRP and TP, and sediment samples for suspended sediment, TP and BAP (bioavailable phosphorous).
Results are reported by McDowell and Wilcock (2004). Levels of phosphate were highest in the water column and in sediment during summer and autumn, and lowest in winter and spring, and generally increase downstream. Sediment entering the stream is sourced primarily from topsoil, and enters via tile drains or, to a lesser extent, overland flow. Direct drainage via mole drains of irrigated dairy effluent contributes one third of the total phosphorous loading to the stream. Phosphorous is considered to enter primarily from the same sources, and exceeds levels desirable for good water quality. It was suggested that management focus on reducing sources of phosphate entering tile drains, primarily by reducing soil Olsen P levels. Levels of E. coli are high, regularly exceeding contact recreation guidelines.
In spring 2001, as in the other best practice catchments, soil physical, chemical and biological quality assessments were made of the major soil types within the catchment. Although overall soil quality was good, about a third of the soils in the Bog Burn were compacted, and Olsen P, which averaged 42, was higher than the target of 25-35.
This work was repeated in winter 2003. At this time soil compaction was at a moderate level, and Olsen P was high on some properties, exceeding 40 in 43 percent of the properties in the catchment. Many sites had lower than optimum soil potassium.
