Managing input of agricultural phosphorus to Oyster Harbour

Key findings

• Agricultural land used for sheep and cattle grazing contributes most phosphorus to waterways because it is the most extensive landuse in the catchment.
• Much of the phosphorus loss is associated with larger or more intense rainfall events and subsequent river flow.
• Phosphorus can be lost from legacy stocks in catchment soil in a dissolved form independent of fertiliser application.
• Phosphorus lost from catchment soil in a dissolved form can be transformed into a particulate form in streams and rivers when combined with soil particulates from stream banks and beds. Nutrient management practices that aim to physically filter particulate nutrients cannot successfully remove dissolved nutrients.
• Phosphorus stratification increases the risk of phosphorus loss when surface runoff dominates.
• The best practice for phosphorus management is to test soil and use critical soil test values to make evidence-based fertiliser decisions.

Background

• Soil type: Swampy soil, sands, loams, clays, gravels
• Average annual rainfall: 400-1000 mm

Agriculture

Much of the Oyster Harbour catchment was cleared for extensive agriculture in the 1950s and 1960s. In the southern parts of the catchment, agriculture primarily revolves around extensive grazing of annual clovers and ryegrass, while cropping becomes more prevalent in the northern areas. Near the coastal regional centres, there are small pockets of intensive agriculture. Pastures serve as grazing land for both cattle and sheep across the region. Fertiliser management practices have been maintained since the land was initially cleared, involving annual application of fertilisers to pastures situated on naturally-infertile soil.

Condition of Oyster Harbour

Mapping of seagrass extent and density in Oyster Harbour identified a 90% loss between 1962 and 1992. Seagrass restricts erosion of sediment and provides shelter and habitat for fish breeding, as well as supporting a rich ecosystem. Growth of macroalgae due to nutrient input from agriculture and subsequent shading of seagrass was identified as the root cause. Phosphorus was identified as the limiting nutrient for algae growth.

Methods

The methods used to identify nutrient sources, transport pathways and their management are based in a range of disciplines including hydrology, soil science, plant nutrition, chemistry, behavioural and data science. In brief, some of the methods and data applied to this case study included the use of discharge measurement structures at gauging stations coupled with a water quality monitoring program, surveys of farmer fertiliser use and timing, soil test data from routine sampling programs and experiments, and planned experiments to test the impact of best management practices on water quality.

Results and discussion

Source and transport factors

A range of factors, broadly classed as source and transport factors influence the delivery of nutrients to waterways. When there is both a source of nutrients and potential for its transport, there is a risk of nutrient loss. Some examples for individual factors include:

• The risk of phosphorus loss increases as soil test phosphorus values increase.
• The risk of phosphorus loss increases when fertiliser is applied at the break of the season compared to spring fertiliser application.
• The risk of phosphorus loss increases as soil phosphorus retention (sorption) decreases.
• Nutrient loss risk increases with increasing rainfall intensity and amount.
• The risk of nutrient impact posed by an enterprise is greater the closer it is to sensitive waterways.

Water quality monitoring

Water quality monitoring of the main riverine input to Oyster Harbour (Kalgan River) shows that the load of phosphorus is strongly influenced by river flow, which is influenced by rainfall and antecedent catchment soil moisture. In a fairly dry year until August, a rainfall event of 60 millimetres produced small river flow and a commensurate increase in phosphorus load. This rainfall event was sufficiently large to increase catchment antecedent moisture so that subsequent smaller rainfall events led to substantial river flow and phosphorus loads later in the year.

Rainfall and river flow had more influence over phosphorus load than the timing of phosphorus application, most of which (75%) occurred from March to May, or the traditional seasonal break. In this example, the disparity between phosphorus discharge and timing means that phosphorus loss is not directly from fertiliser, rather from legacy phosphorus stocks in the soil.

Partitioning of the phosphorus load into particulate and dissolved (soluble) fractions is often used to identify mechanisms through which phosphorus is mobilised into streams and rivers. In 1992, 71% of the phosphorus was transported attached to particulates, and 29% in a soluble form. This strongly suggests that erosion of surface soil containing phosphorus as the primary mechanism of mobilisation. However, hillslope runoff and leaching studies, and small catchment studies in the same catchment indicate that 96–99% of phosphorus lost from hillslopes was in a soluble form, and 2 to 3 orders of magnitude more phosphorus was transported via leaching pathways than surface runoff. This suggests that dissolved phosphorus lost from agricultural landscapes is transformed into particulate phosphorus within streams and rivers. This transformation likely occurs through adsorption of dissolved phosphorus onto particulate matter that is derived from stream banks and beds, rather than from erosion of previously fertilised surface soil.

The implications of these findings are that management practices such as riparian buffers that act to physically filter surface derived particulates laterally to streams are unlikely to be suitable in catchments with sandy soil where leaching is the dominant transport pathway. Small scale experimental work in catchments with sandy soil in south-west Western Australia support these findings, and show that riparian buffers reduce phosphorus transport by <5%, but do reduce suspended sediment transport by 90% through stabilisation of stream banks and beds. Whilst suspended sediment is reduced, soluble phosphorus that is transported to streams from hillslopes via leaching and subsurface transport pathways has much less particulate matter to interact with in streams and is easily transported downstream in a bio-available form.

Soil testing

Soil test records from a survey of 4600 samples (0–10 cm) in the Oyster Harbour catchment in 1987 and 1988 showed that 56% of samples had high phosphorus status. That is, pastures would not respond to applications of phosphorus on these soils as the concentrations of Colwell phosphorus exceeded critical or benchmark values. More recent surveys of 3395 samples collected from 2012–2023 show similar levels of high phosphorus status (62%), along with 40% of samples with potassium deficiency, 30% with sulfur deficiency and over 90% of samples with soil acidity problems. The high phosphorus status soil contained on average 1.9 times as much phosphorus as was required to achieve optimal pasture production. This increases the risk of phosphorus loss by an average of 1.9 times. This risk is increased further when other constraints such as low potassium, low sulfur, or soil acidity limit the uptake of phosphorus already in the soil, and further again if phosphorus is applied when it is not required.

This illustrates the importance of using soil testing to identify constraints to pasture production. Soil test results should then be appropriately interpreted against critical values for pastures or crops (see the box ‘Critical soil test levels’, page XX)In the case of the Oyster Harbour catchment, 62% of sampled paddocks do not require an application of phosphorus to achieve maximum pasture production. Phosphorus budgeted for high phosphorus status paddocks can be redirected to other limiting constraints and will increase the uptake of phosphorus by pastures. Cessation of phosphorus application will gradually reduce soil phosphorus levels, and phosphorus loss risk. Soil testing can identify when critical Colwell-phosphorus levels have been reached, and when maintenance phosphorus applications can commence.

Phosphorus loss risk can be significantly greater than is indicated by soil testing of 0–10 centimetre samples. This is because phosphorus becomes stratified in the soil under pasture systems where fertiliser is surface applied. The result is that phosphorus in the 0–1 centimetre layer of soil is 3 to 5 times more enriched than is measured in a 0–10 centimetre sample. High intensity rainfall or surface runoff due to saturation excess flow will interact with layers of soil that are much more phosphorus enriched, leading to significant surface loss of dissolved phosphorus.

Acknowledgements

This work is a summary of activities largely undertaken by the nutrient management group within DPIRD/DAFWA since 1988, funded by the WA state and federal government . Contributing projects include: South Coast Estuaries Project (WA State Government), National Riparian Zone Program (Land and Water Resources Research and Development Corporation), National Heritage Trust, Whole Farm Nutrient Mapping (WA State NRM Program), Regional Estuaries Initiative (Royalties for Regions), Healthy Estuaries Western Australia (Royalties for Regions).