PHOSPHORUS LOSSES THROUGH TRANSFER, SOIL

EROSION AND RUNOFF:

PROCESSES AND IMPLICATIONS

Ward, J.e. Talbot, J.M. Denne, T. and

Abrahamson, 1>1.

December 1985

Information Paper No.3

Centre for Resource Management

Lincoln College and University of Canterbury

TABLE OF CONTENTS

FOREWARD

1. INTRODUCTION

1.1 The nature and scope of the problem

1.2 Specific objectives

2. PHOSPHORUS LOSS PROCESSES

2.1 Stock transfer

2.2 Management of transfer loss

Page

i

1

1

1

1

2

4

2.3 Water runoff and soil erosion 5

2.4 Management of losses caused by runoff and erosion 15

2.5 Riparian zones and wetlands 16

3. IMPLICATIONS OF PHOSPHORUS IN WATER RESOURCES

3.1 Sources of phosphorus loading

3.2 Effects of phosphorus ·additions on primary

productivity and sediments

3.3 The trophic status of New Zealand lakes

4. CONCLUSIONS AND POLICY IMPLICATIONS

4.1 Future risks

4.2 Management needs and approaches

ACKNOWLEDGEMENTS

REFERENCES

19

19

20

24

29

30

31

33

34

FOREWORD

Phosphate rock is a strategic material upon which pastoral agriculture

and all New Zealanders depend. Phosphate fertilizer has no close

substitute, and is, therefore an important limiting factor to agricultural

productivity. The future wellbeing of the country depends on its

efficient acquisition, manufacture, distribution and use.

This report is part of a larger cross-disciplinary study carried out

by Centre staff on the multiple dimensions of phosphate management in

New Zealand. The report presents an examination of phosphorus losses

from the production system and the attendant consequences on environ­

mental quality. Special emphasis is given to hill country loss

mechanism, where it is known that significant amounts of phosphate

fertilizer are picked up in surface runoff and/or are displaced by

grazing animals. The longer term consequences of nutrient loading

on downstream water bodies are discussed, and the policy and management

implications for maintaining current water quality levels are highlighted.

K.L. Leathers

Study Team Leader

1

1. INTRODUCTION

Phosphorus is added to New Zealand soils to increase plant growth

but is lost from the pasture system by stock transfer, soil erosion

and water runoff. When this nutrient reaches waterways, growth of

aquatic plants is promoted; excessive growth results in changes to

water quality and potential uses of the water. Phosphorus is

retained by lake sediments so additions have long term consequences

on lake and downstream water quality. While phosphorus losses to

the farmer may be small, the costs borne by society may be large

because of the long term effects and the expense of any remedial

measures.

1.1 The nature and scope of the problem

The amount of phosphate fertiliser applied annually to New Zealand

pastures has been approximately two million tonnes in recent years.

To sustain growth of agricultural exports in coming decades, greater

levels of fertiliser phosphorus may be required. Since a significant

proportion of the applied fertiliser is lost to the production system,

efficiency in on-farm use if becoming an important concern for farm

managers and policy makers. On-farm losses raise production costs

and lower farm incomes, while phosphorus enrichment of freshwater

bodies has implications for environmental quality.

1.2 Specific objectives

This paper examines what is currently known about fertiliser phosphorus

losses in New Zealand pastoral systems. It focuses on the potential

above ground losses (and costs) to the farmer, management options

for conserving phosphorus, and on the ecological effects and potential

social costs of fertiliser phosphorus additions to waterways and

approaches to avoiding or managing these unintended impacts. Our

understanding of the issues, and factual evidence which underlines

these two main theses, is explored in some detail.

2. PHOSPHORUS LOSS PROCESSES

Fertiliser phosphorus losses result from four basic mechanisms:

1. below ground phosphorus cycling processes (discussed by

Scott, 1985);

2

2. embodied phosphorus in animal carcases and products

removed from the farm;

3. 'animal transfer' or the displacement of dung phosphorus

to locations on the farm which effectively removes this resource

from entering into subsequent production cycles; and,

4. phosphorus lost via water runoff and soil erosion.

The relative order of magnitude of these losses is not known at present,

nor are the potential savings that might be realised in their control ­

the value of phosphorus conservation. Because the first two mechanisms

of loss are largely uncontrollable this paper considers only the nature

of stock transfer, soil and runoff losses and their potential for

control through improved farm management.

2.1 Stock transfer

Phosphorus cycles in ecological systems through the action of organisms

at different trophic levels. Herbivores have an important role in

cycling. Because of cultural practices, sheep and cattle are the

dominant herbivores of New Zealand pastoral agricultural systems

(although other herbivores such as horses and deer may be locally

dominant), and management aims to maximise their production. Phosphorus (P)

is applied to pastoral systems to increase plant production to provide

food for livestock. Phosphorus is consumed with the plants; some is

retained by the animal and some is returned rapidly to the soil-plant system

via excretion. Although the amount returned may be quite large,

phosphorus is concentrated largely in dung and deposits may be dis-

tributed in such a way that the phosphorus is unavailable to much

of that system.

The distribution of cattle dung is more or less random. However,

O'Connor (1981) found that the annual return of dung at 3 cows ha- l would

cover only 5.5% of the grazed area. This would leave appreciable

areas of the paddock without phosphate from that source for several

years.

3

Sheep dung tends to be clumped on small areas. This is related to

a behavioural characteristic of sheep in which they tend to collect

together at night. Gillingham and During (1973) found a net gain

of 28.6 kg P ha- l on a sheep camp occupying 6.4% of a block and an

average net loss of 5 kg P ha- l on the other areas. Gillingham (1980)

found more than 88% of the dung on campsites in two paddocks of north

and south aspects, these sites occupying 20.1% and 12.2% of the total

pasture respectively.

Areas grazed heavily can be depleted of phosphorus if fertiliser is

not applied (Hilder, 1966). Sheep tend to camp and excrete at high

points, so there is a transference of phosphorus in dung to higher

areas as shown in Figure 1. However, greater movement of phosphorus

in runoff means complete depletion would not occur on steep slopes.

E8~

.s::.-6 g'(1)

4 ~CtI...

C)

10

12

-2

....... No.of faeces

11-. Grass length

o

~

.. \. /---/ "".:::><:~ /.--.

··1........ ...•.......•.... .....•

-E,-12(1)0-'-(1)10-(1)

E~ 8"0

Eo 6

('II

1\en(1) 4­o(1)~­_ 2ooz O ..........-....-......- .........,-....-..,--po-...,.-..,-... 0

6 12 18 24Distance downslope (m)

FIGURE 1: Mean number of faeces> 2 cm diameter per m2

at 3 m intervals down the Wool shed Paddock,Taita, for Enclosure 1 (18 Sept 1975) andEnclosure 2 (20 Oct 1975). Faeces were countedafter grazing; grass lengths were measuredbefore grazing.

(McColl and Gibson, 1979)

4

Overall estimates of phosphorus loss through animal transfer are not

available.

2.2 Management of transfer loss

Reduction of phosphorus loss caused by stock transfer through direct

stock management (shepherding) is not presently practised in New zealand.

Current management practices such as fencing, rotational and controlled

grazing may help.

Gillingham and During (1973) suggest it is possible to avoid topdressing

10% of a block (i.e. stock camps and ridges) without affecting pasture

yields in high country development. However, the economic viability

of this practice has not been demonstrated (Saunders et al., 1981).

In the high country, fencing steep slopes from gentle slopes would

theoretically reduce transfer losses, as sheep would not be able to

move to and camp on the easier slopes. In practice, the complexity

of land topography limits this.

Planting trees in pasture is a potential source of phosphorus conservation.

Recent experiments by M. Belton and K.F. O'Connor (in prep.) show

that Olsen P levels (plant available P) in soils of different types

are three times higher under exotic forest than under unimproved or

improved pasture. By spacing the trees, sheep would be kept dispersed

when they seek shade or shelter and slope stability would be enhanced.

Systems of management may affect the distribution of dung. While

neither Gillingham (1982) nor Thorrold et al. (1985) could find discernible

differences between continuous and rotational grazing on dung distribution,

stocking rates may influence dung distribution. Thorrold et al.,

(1985) have found that a more even distribution of dung was obtained

on slopes by high stocking rates. Daytime dung frequency was found

to be more closely related to grazing distribution than to resting

or total animal distribution. Correlations between soil Olsen P levels

and dung frequency were low even when night camp areas were included,

but Olsen P levels were consistently higher in paddocks with higher

stocking rates. It is suggested (ibid) that at high stocking rates,

pasture consumption is high and a more even distribution of phosphorus

5

is returned to the soil as dung, perhaps in more quickly released

forms. At low stocking rates, pasture consumption is low and much

of the phosphorus is retained in the herbage or in the litter from

which it is released more slowly.

There are no quantitative studies on the relative turnover rates of

dung and litter in New Zealand pastures. Microorganisms, earthworms,

temperature and moisture are some of the factors involved in this

process. Addition of earthworms is an effective and increasingly

practiced method of increasing litter and dung turnover rates (Stockdill

and Cossens, 1984), especially in the earthworm deficient yellow brown

earths of otago and Southland.

The economic viability of many indirect methods of stock management

to reduce transfer loss of phosphorus are not known. Increased use

of some methods, such as silvipastoralism and selective topdressing,

appear warranted for the other benefits that accrue.

2.3 Water runoff and soil erosion

Phosphorus is present in soils in various forms which vary in solubility,

and consequently in susceptibility to movement by water. Phosphorus

can be lost from soil by dissolution when contact is made with water

or by erosion of soil particles containing phosphorus. Erosion may

occur by wind or by water. Lost phosphorus may be transported to

waterways and aquatic biological systems in runoff water and attached

to soil.

Phosphorus losses through soil erosion by wind and water are variable

in New Zealand and are related to the land form, soil type, intensity

of land use and climate. In the high country, losses through soil

erosion may be high due to slope instability and heavy rainfall.

In Central Otago and the Canterbury plains wind erosion causes signifi­

cant phosphorus losses. The few rates of both wind and water soil

erosion that have been measured point to very high soil losses by

overseas standards on many New Zealand farms (Painter, 1978). Phosphorus

losses through wind erosion have not been quantified but, as with

runoff, may be high on cultivated or recently fertilized land.

6

On well-managed pastoral land phosphorus losses in runoff are usually small.

Because of the reactivity of phosphorus in soils these amounts are

generally regarded as insignificant from an agricultural perspective.

But as phosphorus is often a limiting nutrient for aquatic biological

systems, small additions of phosphorus may have a considerable effect

on the biological productivity of rivers and lakes.

Runoff may be categorised as surface, subsurface or groundwater as

defined by Langbein and Iseri (1960 in Ryden et al, 1973). These

different types of runoff transport different forms of phosphorus.

Surface runoff transports particulate and dissolved phosphorus, sub­

surface runoff transports dissolved and some colloidal phosphorus,

and groundwater transports dissolved phosphorus (Ryden and Syers,

1973).

There have been some differences of opinion in the New Zealand literature

regarding the quantitative importance of different runoff types for

the transport of phosphorus. It is widely recognised that phosphorus

is strongly held by most soils, and that losses in subsurface and

groundwater runoff are insignificant in comparison with losses in

surface runoff (Baker et al., 1975). But some workers have suggested

that fluctuations in available phosphorus in subsurface runoff may

often be similar and in some cases higher than those in surface flows

(Ryden et al., 1973; Syers, 1974). While it can be argued that the

soils tested by these authors are prone to leaching, it is still apparent

that there is uncertainty regarding the importance of different runoff

types to water enrichment.

Table 1 presents various data on the amount of phosphorus (loading)

in different forms of runoff from pasture systems. The largest amounts

of all forms of phosphorus appear to be carried in surface runoff.

Particulate phosphorus (PP) incorporated in soil particles comprises

the major portion of total phosphorus (TP) carried. Thus Burwell

et al., (1975) found 96% of TP was transported in the form of PP over

a variety of soil cover treatments. Owing to the low energy of surface

runoff there tends to be preferential transportation of small sized

particles (clay and silt sized colloids). Small soil particles tend

7

to contain higher amounts of phosphorus than large particles. Thus

eroded material has higher concentrations of phosphorus than the soil

from which it is derived (Stoltenberg and White, 1953).

TABLE 1: Phosphorus loadings in runoff

Runoff Form P 10a?-ing1

P Fertiliser ReferenceComponent ofP* kg ha yr- applied

N. AmericaGroundwater TP 0.1 + Minshall et a1.

(1969 )

Subsurface TDP 0.01 Bolton et a1.0.12 + (1970)

Surface and TP 0.7 § + Witzel et a1.subsurface (1969)

Surface Soluble P 0.22 + Schuman et a1.TP 0.28 + (1973)

Surface Available Stoltenberg &

P 0.2 + White (1953)

New ZealandAccelerated DIP 0.07 - 0.44 + Sharpley and

subsurface TDP 0.10 - 0.52 + Syers (1979c)TP 0.18 - 0.89

Surface DIP 0.28 - 0.50 Sharpley & Syers1.66 - 2.80 + (1979c)

TDP 0.31 - 0.611.66 - 2.89 +

TP 0.85 - 1.283.67 - 5.63 +

Surface TP 0.7 - 1.0 + Lambert et a1.(1986)

*TP = Total Phosphorus; DIP = Dissolved Inorganic Phosphorus;TDP = Total Dissolved Phosphorus

~Based on extrapolation assuming direct relationship betweenrunoff and nutrient loss.

As additions of phosphorus to waterways via surface runoff generally

occur in large events, phosphorus concentrations in receiving waters

8

may be greatly increased over the short period of time. However,

the immediate effects on the receiving streams may be minimal as amounts

of phosphorus available may be greater than that which can be used

in primary production (McColl et al., 1977). Groundwater and subsurface

runoff provide a more constant supply of phosphorus to waterways.

The effects of phosphorus from these sources on water quality may

be greater than their smaller phosphorus loading suggests when compared

to the effects of phosphorus from surface runoff.

The effects of inputs of phosphorus on receiving waters may be limited

by the absorptive capacities of the sediments. Much of that arriving

in runoff may quickly react or become absorbed. Conversely a very

large proportion of the phosphorus concentration of waterways derives

from stream-bank erosion and the release of inorganic phosphorus (IP)

from suspended and in situ particulate material (Sharpley and Syers,

1979c). Sharpley and Syers (ibid.) present data which indicate

that the major proportion of PP (83%) and TP (71%) transported annually

in stream flow is derived from stream-bank and stream-bed material.

The contribution of dissolved inorganic phosphorus (DIP) and total

dissolved phosphorus (TDP) from these sources were 27% and 42% respectively.

Stream sediments appear to function collectively as a slow-release

pool of phosphorus. Phosphorus which is rapidly absorbed when it

enters the stream is made available over time. Thus the immediate

effects of additions of phosphorus to streams are reduced, but the

effects extend over longer periods of time.

Absorption-desorption processes mean phosphorus loadings can be quite

different at different locations in a waterway or catchment. Measurements

of high phosphorus losses from small plots may not be evidence for pro­

portionally large losses from whole catchments. For example, McColl·

(1978b) measured concentrations of reactive phosphorus in a stream3

as high as 13.4 g m- after the application of fertiliser: losses

of reactive phosphorus from the subcatchment were up to 1 kg ha- l

in a single storm: however, losses of reactive phosphorus from the

whole basin averaged 0.004 kg ha- l .

9

Effects of fertiliser on loading

Different types of phosphorus fertiliser probably produce different

effects in terms of additions of phosphorus to waterways. Sharpley

et al., (1978) conducted experiments with dicalcium phosphate (DCP)

as a less soluble alternative to monocalcium phosphate (MCP, as

superphosphate). Both were applied at levels of 50 kg P ha-l . Although

in the short term greater amounts of phosphorus were transported in

surface runoff from MCP applications, the annual TP loss from areas

with DCP applied was 7.09 kg ha-lvs. 5.63 kg ha- l from MCP, equivalent

to 11.5% and 8.8% of fertilizer phosphorus respectively. To our

knowledge, other fertilisers have not been examined for their runoff

effects.

Applications of fertiliser clearly increase the loss of phosphorus

from agricultural land. But some of the dramatic increases in phosphorus

concentrations are measured immediately after fertiliser applications

and are the result of fertiliser falling directly into streams. The

physical and biological effects of these additions depend on sorption

reactions within those streams and dilution within the whole catchment.

In experiments by Sharpley and Syers (1979c) 50 kg P ha- l yr- l was

applied to one of duplicate plots in June of three consective years

and phosphorus losses were measured over the whole study period.

Comparison of phosphorus losses in surface runoff from undrained,

fertilized and unfertilized plots indicate that between 2.9 and 4.8%

TDP and 5.7 and 8.8% TP was lost annually. These losses of fertilizer

phosphorus were greater than observed in overseas studies (Table 2)

although shorter time periods were measured in the latter. Large

losses of phosphorus (6%) were also recorded in runoff from fertilized

pakihi soils on the West Coast of the South Island (Lee et al., 1979).

McColl (1978b) recorded fertilizer phosphorus losses from 0.02 to

1.9% (Table 2). It is suggested (ibid.) that this range in losses and the

low values may be due to chemical adsorption during downstream transport

and hydrological differences in the stream channel regime due to

riparian vegetation. A conservative estimate of 1-2% of applied fertiliser

phosphorus lost in runoff (Table 2) would represent a total of approximately

1600-3200 tonnes P y-l (assuming 1-2% of 2 million tonnes superphosphate

y-l with 8% P).

10

TABLE 2: Percentage losses of P forms from agricultural landafter fertiliser applications

Form of P % fert.P lost

Soluble P 1.01.2

Soluble P 1.3

Soluble P 0.6TP 0.7

DIP 2.0PP 2.9TP 5.3

TDP 2.9 - 4.8TP 5.7 - 8.8

TP 6.0

TP 0.02 - 1.9

TP 2.9

Fertiliser Length Sourcerate 1 of study(kg P ha- ) period

U.S.A.56 Nelson and Romkens

112 (1970 in Burwellet al. 1975)--

50 3 mths Nelson and Romkens(1971 in Sharpleyand Syers, 1979c)

39 6 mths Schumann et a1.(1973)

New Zealand30 4 mths Sharpley and Syers

(1979b)

50 3 yrs Sharpley and Syers(1979c)

42 5 mths Lee et al (1979 )

60 - 76 2~yrs McColl (1978b)

38 4 yrs Lambert et a1. (1985)

Although the percentage of fertiliser lost in runoff (Table 2) is

usually insignificant to the agricultural system (McColl, 1978b),

the change in phosphorus concentrations in receiving waters may be

considerable. For example, the Waingaehe stream which flows into

Lake Rotorua, normally has an orthophosphate concentration of around

Topdressing of adjacent land caused a four-week pulse of

orthophosphate which peaked at 890 mg m- (Fish, 1969), one hundred

times the natural level. Similar trends in change in orthophosphate

concentrations were recorded by Mitchell (1971), Sharpley and Syers

(1979b, 1979c) and Turner et al., (1979).

Effects of animals on loading

Animals can be responsible for large additions of phosphorus to runoff

waters. They exert three major effects on the soil-grass system (Cooke,

1981): by eating the grass short the soil surface is more susceptible

to erosion; closely grazed grass cannot sieve particulate matter in

11

the manner that long grass can; and animal trampling lowers the

infiltration capacity of the soil and increases the likelihood of

surface runoff. In addition, grazing animals are more directly

responsible for additions of phosphorus through dung. Data relating

to the increased concentration of phosphorus in runoff attributable

to grazing are presented by Lambert et al., (1985) and by McColl and

Gibson (1979), Sharpley and Syers (1976a, 1979a) and Turner et al.,

(1979).

Sheep

Scott (1985) has shown that the amount of P returned as sheep dung

(for an 18 stock unit ha-l grazing system) was about 26 kg ha- l yr- l

or equivalent to 0.33 t of superphosphate. Transfer of some of this

dung to waterways is a potential source of phosphorus loading. McColl

(1978a) estimated that dung provided a greater amount of total P to

Lake Tutira than any other source (Table 3).

TABLE 3: Estimated phosphorus loading on Lake Tutira (catchmentsources)

Input

Fertiliser runoff (assuming 1% loss)

Dung/urine runoff

Subsurface runoff

Soil erosion

Other sources

TOTAL

Source: after McColl (1978a)

Cattle

Phosphorus

700 kg y-l

300-1600 kg y-l

300 kg y-l

1000 kg y-l

100 kg y-l

2400-3700 kg y-l {14-21 kg ha- l y

In contrast to sheep who tend to remain on the margins of waterways

and wetlands, cattle are often observed wading in bogs, ponds and

12

streams. In dry seasons cattle are often grazed on swamplands. This

may result in stream channel widening and a change in water column

structure (Skovlin, 1984) and modification of the land drainage system

by interrupting stream flow (Hughes et al., 1971). The cumulative

impact on the catchment of fluctuating stream levels may be considerable,

particularly in downstream locations. Cattle access to these areas

should be strictly controlled to maintain drainage. The advantages

of an intact riparian zone in reducing nutrient losses (see p.16)

and drain clearance costs need to be made clearer. Controlled grazing

by livestock need not affect the ability of the riparian zone to

filter nutrients (Skovlin, 1984). However, the physical conditions

of this zone must be understood and matched to the grazing regime.

The grazing of cattle can have a large effect on phosphorus levels

in stream flows. Sharpley and Syers (1976a) found the loss of total

phosphorus (principally DIP and PP) in surface runoff due to animals

alone, in the weeks following a 24 hour period of grazing by dairy

cattle (25 ha-l ) to be 0.77 kg P ha- l . (This is 27% of that lost

as a result of fertilizer addition alone.) With two or three grazing

events during the period when runoff occurs in the study area (June-1

to September) a maximum of 5.1 kg P ha could have been lost in surface

flow as a result of both cattle grazing and fertilizer addition (10%

of that added). In a later study, Sharpley and Syers (1979b) found

that ten hours of grazing at a pressure of 25 cattle ha- l resulted

in a 100-fold increase in both PP and sediment in a nearby stream.

The effect was short-lived, however, and concentrations decreased

to the pre-grazing levels within two days. The observed increases

could be attributed to the movement of cattle in the stream channel

stirring up bottom sediments and depositing dung in the stream.

Lambort et al., (1985) report runoff losses of 1.5 kg P ha-l y-l

from catchments with rotationally grazed cattle as opposed to losses

of 0.7 kg P ha-l y-l from sheep grazed catchments.

Earthworms

In addition to the effects of grazing animals, earthworms can exert

some influence on phosphorus losses in runoff. Surface casts of

earthworms contain a higher proportion of fine particles than underlying

13

soil and appreciably greater concentrations of inorganic and organic

P. Casts have been found to be more readily transportable in surface

runoff than soil (Sharpley and Syers, 1976b). These authors estimated

for one study area that approximately 14 kg ha- l of inorganic P and

11 kg ha-l of organic P were accumulated in casts during one year.

They also found that casts released considerable quantities of inorganic

and organic P to solution.

Effects of land use on loading

Different land uses produce different levels of runoff. Ryden et

ale (1973) present data relating to runoff from land under different

crops; highest nutrient losses occurred from vegetables, corn, soybeans

and wheat. Losses of phosphorus from intact forest watersheds are

generally much smaller than from agricultural land (Berg, 1980; Cooke,

1979, 1981; McColl et al., 1977; Ryden, et al., 1973; Syers and Ryden,

1973). Cooke (1981) suggests that typical exports of phosphorus from

forest catchments are approximately one tenth of those from pasture catch­

ments. Ryden et al., (1973) found runoff from unfertilised forest

land contained concentrations of P lower than that of incident rainfall.

Therefore forests conserve phosphorus. Surface runoff from forests

is also low owing to the protection provided by both canopy and forest

floor vegetation, and the high infiltration capacity usually found

at the ground surface.

Owing to the major contribution to runoff made by phosphorus carried

by eroded sediments, together with the value of vegetation in reducing

runoff and sediment movement, losses from bare ground may be consider­

able under any circumstances. Several workers cited by McColl and

Syers (1981) found significant correlations between percent bare ground

and quantities of surface runoff.

Phosphorus in urban runoff originates from sewage, stormwater, industrial

effluents and other waste discharges. Total discharges of phosphorus

in sewage effluent from urban areas in New Zealand have been estimated

at approximately 9000 tonnes y-l (Syers, 1974). Polyphosphates, which

are very soluble and biologically active ingredients of many detergents

(especially in the dairy industry), are significant sources of phosphorus

to waterways (OECD, 1981). Since most urban areas in New Zealand

14

are coastal, total discharges of phosphates to freshwater systems

are probably smaller than those to the sea. McColl (1982) estimated

that 1500 tonnes y-l phosphorus from sewage is discharged to freshwaters.

Preliminary estimates of phosphorus contributions to freshwater systems

from various sources are reported in Table 4. Point source discharges

are apparently only slightly greater than diffuse sources: 18 and

15 tonnes of phosphorus per day respectively. However, fertiliser

phosphorus ranks along with the dairy industry as the highest source

of phosphorus input. Diffuse source dung is next in relative importance.

TABLE 4: Estimates of total phosphorus entering New Zealandfreshwaters

Type of Discharge

POINT SOURCES

Dairy industry

Meat industry

Human sewage

Cowsheds

Piggeries

TOTAL

DIFFUSE SOURCES

Dung

Fertiliser

TOTAL

Source: McColl (1982)

Amount of P (tonnes day-I)

9

0.3

4

1

3.5

18

6

9

15

Point source discharges tend to be localized in certain areas of

the country, particularly in the North Island and near urban centres.

Diffuse sources of phosphorus are more widespread, although they

too are higher in regions of intensive land use. North Island soils

receive two thirds of the applied fertiliser in New Zealand. In 1981-82

approximately 2,000,000 tonnes of phosphate fertiliser was produced.

15

1,347,000 tonnes (68%) was used in the North Island and 624,000 tonnes

(32%) was used in the South Island (New Zealand Meat and Wool Board,

1985). Table 5 shows the breakdown of phosphatic fertiliser into

upland and lowland regions during 1981-82.

TABLE 5: Distribution of phosphatic fertiliser use in New Zealand,1981-82

Location tonnesfertiliser

%

North Island

South Island

Upland

Lowland

Upland

Lowland

457,000

890,000

102,000

522,000

23.2

45.1

5.2

26.5

Source: data from N.Z. Meat and Wood Board (1984)

The consequence of this discrepancy between the islands is clearly

seen in lake trophic levels in the North and South Islands (Figures

3a & b, P 25 & 26).

2.4 Management of losses caused by runoff and erosion

Farm management practices leading to improved water quality,

by reducing fertiliser losses in runoff and erosion, can also reduce

fertiliser requirements without appreciable loss in production or

profit. Such practices include:

- well-engineered roads and tracks, including use of vegetated

buffer strips on downhill slopes;

- the use of well-constructed shelter belts to reduce wind erosion

transport of soil to water bodies;

- improvement in the application of fertiliser - applying the

correct amount at the right time to the land types capable of

generating the highest relative return to the farmer;

- reducing stocking rates in areas prone to significant runoff

problems;

16

- avoiding direct application of fertiliser to stream channels,

seepage zones, land adjacent to lakes and streams, stock camps

and well-defined ridge tops;

- maintaining longer grass at the time of fertiliser application;

- practising conservation tillage, especially overdrilling in

hill and high country development;

- utilising trees more fully, e.g. planting fast-growing thin­

crowned species such as poplars in gullies and on steep hillsides

to control erosion or use as cattle fodder, growing a tree

crop in conjunction with pastoralism (silvipastoralism);

- improving overall grazing management, e.g. fencing or correct

stock rotation; and

- improved drainage of land e.g. mole drains (Sharpley and Syers,

1976).

The relative and absolute value of each of these practices in specific

situations is at present not known. More research is needed on these

to enable the best choice of management options to be applied in a

given situation.

2.5 Riparian zones and wetlands

Riparian zone management offers the last chance to filter out nutrients

and sediment before they enter waterways (Pittmans (unpublished),

1979; McColl and Hughes, 1981; Williams and Brickell, 1983). Wetlands,

and areas away from waterways but prone to surface runoff because

of the poor soil infiltration capacity, may also be managed similarly

(Williams and Brickell, 1983; McColl, 1983).

As water passes through the riparian zone suspended matter is filtered

by the soil (McColl and Syers, 1981) or, in wetlands, the particles

settle out, trapping absorbed and organically bound phosphorus in

the sediment (Klopatek, 1978). Dissolved inorganic phosphorus (DIP)

remaining in the water, in solution in the soil or sediment (interstitial

water), or released from the sediments if conditions become anaerobic

(Richardson et al., 1978) is taken up by micro-organisms and plants.

For rooting plants the depth of the rooting zone and thus the species

composition influences the effectiveness of interstitial

17

phosphorus utilization (Williams and Brickell, 1983). Species composition

is also important for year-round uptake of phosphorus. Few overseas

studies have looked at these two aspects (Von Oertzen, 1981), and

no studies have been undertaken in New Zealand.

The efficiency of the riparian zone as a phosphorus trap is affected

by the hydrological regime which may vary from year to year (Sloey

et al., 1978; van de Valk et al., 1978), the biological productivity

of the system and the concentration of phosphorus (Klopatek, 1978).

Under conditions of high phosphorus loading the DIP uptake capacity

of plants soon becomes saturated (Richardson, 1985), and although

particulate phorphorus is still trapped the efficiency of the zone

is reduced. Further, riparian vegetation usually has only limited

possibilities for phosphorus storage unless periodic harvesting is

undertaken (Richardson, 1985; Young et al., 1980). Richardson (1985)

found that 35-75% of the phosphorus taken up by plants was released

during winter dieback. Although this may be partly retained in the

soil or sediments, net retention is dictated by precipitation events

during the time of maximum senescence and nutrient release in late

autumn and winter (Prentki et al., 1978).

Adequate water quality management should be able to be achieved by

responsible land management. Management practices include:

- limited or lax grazing of riparian land and exclusion of

stock when runoff risks are high (Yates, 1971; McColl,

1983);

- using riparian land only for meadow hay production or for

growing specialised trees, especially fodder trees which

are periodically harvested;

- leaving a grass border between tilled land and water

(Burwell et al., 1975);

- protecting and encouraging vegetation of stream and lake

sides, especially thick ground cover e.g. leaving a zone of

native vegetation; and

- prevention of stock entry to streams and wetlands

(permanent stock exclusion fencing would be essential only

in sensitive areas).

18

In all cases where harvesting takes place by grazing or mechanical

means, care must be taken to maintain soil permeability by avoiding

compactation and pugging.

Research to test the operation of the natural water treatment process

ascribed to the riparian zone is only just beginning in New Zealand

(Williams and Brickell, 1983). Nevertheless, the limited evidence

available indicates that riparian zones justify more care than most

other parts of a catchment (McColl, 1978b).

19

3. IMPLICATIONS OF PHOSPHORUS IN WATER RESOURCES

Phosphorus is an essential element for plant growth. In water it

is required by phytoplankton, benthic algae and rooted plants which

are basic living components of aquatic ecosystems. When a large amount

of phosphorus is present in the system, excessive plant growth may

occur. This is acceptable in some countries and in some circumstances.

For example, in China green shallow lakes are an asset for fish farming.

In developed countries, eutrophic lakes (meaning those with 'good

food') are often considered unacceptable because their high biological

production conflicts with scenic and recreational values. Large sums

of money have often been spent cleaning up eutrophic lakes.

The purpose of this section is to review the implications of phosphorus

in water, and to discuss the options for managing the use of phosphate

fertiliser with respect to water quality objectives.

3.1 Sources of phosphorus loading

Phosphorus enters freshwater systems in many ways. Firstly, phosphorus

may be carried into rivers and lakes by way of the following:

- water from springs, streams (including transported

sediment), surface runoff and ground water movements;

- air, such as from rainfall, wind-blown soil, fertiliser,

leaves and dung;

- direct contact with the soil along stream and river

banks and areas of swamp or flooded land;

- plant and animal biomass through natural decay and direct

deposition by animals;

- direct discharge from urban and agricultural sources; and

- release from sediment during periods of anoxia or by

catastrophic events.

Secondly, phosphorus loading may be considered as a point source ­

diffuse source continuum. Point sources include, for example, discharges

from dairy and meat industries and treated sewage effluent. Diffuse

sources, on the other hand, include such sources as runoff from agric­

ultural land. Between these classifications exist cases which may

20

fall into either category, such as stockyard runoff or tip leachates.

A similar situation applies to dung movement. Point sources are much

easier to quantify than diffuse sources.

Phosphorus loading may also be viewed from its temporal aspect. Sources

of phosphorus loading may be continuous or periodic. Continuous sources

of phosphorus typically come from such sources as springs, while

urban discharge, stream inflow and surface runoff sources may come

in pulses according to human decisions or climate. Catastrophic events

such as freak storms, major floods and earthquakes are hard to predict

and equally hard to measure. By contrast, continuous sources are

more easily quantified.

3.2 Effects of phosphorus additions on primary productivity and sediments

Primary productivity changes

Any effect that phosphorus has on a water body will depend on the

form it is in when it reaches the water. If it is present as dissolved

inorganic phosphorus it is immediately available for plant growth.

Phytoplankton and floating plants obtain their phosphorus requirements

from the water column, while submerged rooted plants obtain theirs

from the sediments and the water column (Bristow and Whitcombe, 1971;

Denny, 1972). Phosphorus may be locked up in plant biomass both in

streams (Vincent and Downes, 1980) and lakes (Mitchell, 1975) and

released again during algal and macrophyte decay. It is released

more slowly from macrophytes than from phytoplankton, but the annual

decay of macrophytes results in the liberation of significant amounts

of phosphorus. In Lake Rotorua, 7 tonnes of phosphorus are (potentially)

released from macrophytes each year (Richmond, 1978).

Where phosphorus is limittrrg' the effect of fertiliser phosphorus on a

water body is to increase primary productivity. In lakes, this

increase is usually reflected in high phytoplankton numbers (Golterman,

1973; Mitchell, 1975). However, in some lakes increased phosphorus

in the water results in increased biomass of macrophytes (Mitchell,

1971; Haumann and Waite, 1978) or of filamentous algae (Howard-Williams,

1981). The reasons why either phytoplankton or macrophytes show a

dominant response to phosphorus additions in a given situation are

21

unclear but appear to be releated to the availability of light and

the nutrient balance in the water (Mitchell, 1971; Spence, 1982).

Patterson and Brown (1979) suggest that increased phytoplankton prod­

uctivity results in more suspended material in the water. This is

filtered by the macrophyte beds and accumulates as sediment with high

nutrient content (Brown and Dromgoole, 1977). The sediment then

stimulates further weed growth. In this way eutrophication may

accelerate the growth and spread of weed by enabling suitable sub­

strates to be formed more quickly than under low nutrient conditions.

The problems caused by increasing productivity are of several types.

Algal blooms cause loss of water clarity and unsightly scums may result.

Concentrations of oxygen may drop in deeper waters during plankton

blooms due to increased bacterial activity resulting from sedimentation

of dying plant cells. Decreased oxygen levels may in turn affect

fish such as trout living in the deeper lake waters. Blue-green algae

periodically become the dominant algal type under high nutrient

conditions. Some of these are nitrogen fixers which may produce

substances toxic to mammals. The nuisance value of excess macrophytes

in lakes and reservoirs is well known both for clogging drains, streams

and water filters and for interfering with boating, swimming and fishing.

Sediments

The sediments in a water body are a sink for phosphorus. They may

also be a source of phosphorus under some circumstances and are an

active centre for phosphorus cycling. Knowledge of ~hether the sediments

act as a source or sink for phosphorus is essential for predicting

the effects of different phosphorus loadings and for planning lake

restoration (Bostrom et al., 1982).

When phosphorus enters an oligotrophic lake the sediments provide

a sink for a major portion of the inflow (up to 96%, Twinch and Breen,

1978), either by direct binding by the bottom sediments or by sedi­

mentation of suspended material. In lakes with higher nutrient levels,

phosphorus may be released from the sediments and transported to the

water column. This internal loading of phosphorus into lake water

may exceed deposition for certain periods of the year, often increasing

22

the primary productivity of the lake. A detailed discussion of the

phosphorus interactions between the water and sediments is given by

Syers et al., (1973) and Bostrom et al., (1982). Since dissolved

inorganic phosphorus is required for primary productivity, the influence

of other phosphorus compartments in both sediment and water on the

dissolved inorganic phosphorus is important. Figure 2 illustrates

the major phosphorus compartments and the interactions between them.

/Dissolved / Dissolved ,p. P

~ 0

I 1"M~1Particulate "- Particulatep. P~ 0

EX'i'ERNAL 1\ SETTLING, MIXING A INTERNALHOSPHORUS .J SEDJ:.lENT DIFFUSION V PHOSPHORUS

Sediment: SedimentParticulate Particulate

p. p~ 0

~ v;YSediment SedimentDissolved / Dissolved

p. P~ 0

FIGURE 2: Interchange between water and sediment P compartments

(p.= inorganic Pi P = organic Pl.~ 0

(Syers et al., 1973)

Nutrient exchange with the overlying water may occur from up to 20 cm

deep in the sediment. This is also the zone of greatest fungal and

bacterial activity. The phosphorus concentrations near the sediment

surface are very variable and likely to be influenced by many factors

including shape of the lake basin, water movement, water depth and

trophic state. This variability suggests care should be taken when

comparing lakes (McColl, 1977). The deeper sediments may reflect

23

more accurately the sediment characteristics of the lake (Fish and

Andrew, 1980). Sediment cores from Lake Rotorua show total phosphorus

decreases from about 1.5 mg g-l dry sediment at the sediment surface

to an approximately constant rate of about 0.8 mg g-l 20 cm below

the surface (Fish and Andrew, 1980).

The interstitial water between the sediment particles contains less

than 1% of the total phosphorus in the sediments but it is important

because it is the phosphorus directly exchangeable with the lake

water (Syers et al., 1973). The concentration of total dissolved

phosphorus in the interstitial water is much higher (usually 5-20

times) than in the lake water (Bostrom et al., 1982). There is also

a large difference in phosphorus concentration in interstitial water

between oligo- and eutrophic lakes. Unlike the total phosphorus content

of the lake sediments, the interstitial water concentration reflects

the trophic state of a lake (Bostrom et al., 1982).

Once phosphorus has entered the sediments of a lake system, it becomes

a major problem from the point of view of eutrophication control.

Sediments release inorganic phosphorus to the overlying water during

periods of anoxia. The more eutrophic the lake becomes, the more

frequent and severe will be the anoxic conditions of the lake bottom,

and consequently the more phosphorus will be liberated. Phosphate

release from temporarily deoxygenated sediments of eutrophic Lake

Rotorua during a period of summer stratification was estimated as

20-40 mg m-2

d- l . In contrast, the rate at which phosphorus returned

to lake sediments during oxygenated conditions was measured at about

0.5 mg m-2

d- l (White et al., 1978). More recent experiments (Vant,

1985) estimate that 35 tonnes of phosphorus were released from the

sediments of this lake over the 1984-85 summer.

Once eutrophic, lakes will probably remain in that condition for some

time even after all external sources of phosphorus are removed, because

of phosphorus release from the sediments (Syers et al., 1973).

Dredging the lake may reduce the problem, but equally, dredging may

expose lower sediments which then become new sources of phosphorus.

24

Activities that stir up the sediments such as storms or harvesting

of macrophytes may also promote phosphorus release. Reversal of

eutrophication through control of phosphorus input is possible

although a return to the oligotrophic condition would be slow and

costly.

For waters that have become eutrophic, various treatment possibilities

exist:

- mechanical removal of sediments from shallow lakes

or bottom sealing;

- diversion of phosphorus rich waters entering lakes

or reduction in input concentration;

- selective discharge of nutrient enriched bottom water;

- artificial circulation or aeration to maintain oxygenated

bottom water;

- reduction of water retention time or flushing with water

of low nutrient content;

- chemical treatment with Iron or Aluminium compounds to remove

dissolved and particulate phosphorus and increase

phosphorus retention by the sediments (Syers et al., 1973); and

- harvesting of aquatic weeds.

Most of these management techniques have been used in New Zealand.

For example, dredging of sediments has been used in Tomahawk Lagoon

(Mitchell, 1971). In Lake Tutira some inflow water has been diverted

and the lake water has been artificially circulated by air from

October to March to maintain oxygenated bottom water (Tierney, 1980).

3.3 The trophic status of New Zealand lakes

It is useful to classify lakes according to the amount of primary

productivity. The productivity of a lake depends on the nutrient

input, the size of the lake basin, the residence time of the water

in the lake and phosphorus retention by the sediments. Oligotrophic

lakes have low nutrient levels and relatively low primary productivity

Most lowland lakes in New Zealand are classified as eutrophic and

most upland North Island lakes are also tending towards eutrophic

status (Figure 3a & 3b). Upland South Island lakes so far have

25

//

Legend

o 01 igotrophic

() Mesotrophic

• Eutrophic

FIGURE 3a: Trophic status of North Island lakes

o

o..0

oo0o

oo

~

o

8eo

9<:

27

relatively low trophic status, with a few exceptions such as Lake

Hayes, Lake Johnson and Lake Alexandrina (Figure 3b). Apart from

a few urban catchments, the trophic state of New Zealand lakes

reflects the intensity of agricultural development in the relevant

catchments (White, 1982). In the South Island, further development

of irrigation and concommitant increase in land use intensity in many

areas is likely to promote greater movement of nutrients into water

bodies. There is accumulating evidence to suggest that the trophic

status of many upland South Island lakes is changing as greater

development occurs in the high country (Stout 1981, Malthus 1985).

In the Lake Mahinerangi catchment, 30% development over 16 years has

resulted in nine times greater phytoplankton production along with

increased total nitrogen and total phosphorus (Table 7) (Malthus,

1985).

TABLE 7: The relationship between catchment development andphytoplankton production in Lake Mahinerangi andinflow stream nutrient concentration

% lake Phytoplankton % stream Inflow Inflowcatchment production catchment stream streamdevelopment mg C m-

2d- l development Total N Total P

- 3 - 3mg m ~m

1964-66 0 76 0 132 13

1968-70 3 210

1976-78 380

1980-82 30 630

100 451 86

Source: Mitchell and Galland (1981), Malthus (1985)

Phosphorus concentration in water is a commonly measured indicator

of trophic status. However, other factors can also be used to dist­

inguish one trophic state from another. These include water transparency,

chlorophyll a, oxygen and total nitrogen (McColl, 1972, 1977; White,

1983). It is now generally agreed that no single quantitative parameter

can be used to distinguish between trophic states. Table 8 shows

the average total phosphorus concentration in lakes of different trophic

28

status in New Zealand. Data obtained from the OECD programme for

eutrophication research is also reported (White, 1982). The large

range for each category clearly illustrates that total phosphorus

alone is not sufficient to categorise each lake.

Several nutrient loading models have been developed in the USA and

Europe (as part of the OECD programme on eutrophication) to describe

the relationships between nutrient load and trophic condition.

Phosphorus loading curves have been used to estimate the specific

phosphorus loading on a lake necessary to produce a particular trophic

state (Vollenweider and Dillon, 1975). However, some New Zealand

lakes do not fit well on these curves, such as Lake Taupo (White, 1982).

This is because New Zealand lakes exhibit low levels of nitrogen,

compared with lakes sampled in the OECD study, and may be nitrogen

limited at times. Caution is therefore necessary in the use of OECD

predictive equations (White, 1983).

TABLE 8: Average total phosphorus concentration in lakesof different trophic status

MESOTROPHIC EUTROPHIC- mg m-

mg m

Rotoaira 13.7 Ohakuri 29.1

Okareka 14.1 Rotorua 26.0

Rotoiti 21.1 Rerewhakaaitu 15.9

Mahinerangi 23.8 Okaro 56.0

Ngapouri 51.6

Tutira 60.3

Rotongaio 231

Horowhenua 501

OLIGOTROPHIC

Taupo 5.5

Aratiatia 1l.5

Tikitapu 6.4

Rotorua 11.0

Okataina 17.6

Waikaremoana 4.2

Ahaura 8.7

Brunner 16.5

Haupiri 21.5

Hochstetter 10.4

Kangaroo 14.0

Kaniere 13 .0

Lady 8.8

Poerua 14.0

N. Z. lakes 11. 7 mean

OECD study

4.2-21. 5 range

8.6 mean

3.0-16.1 range

20.8 mean

13.7-31. 3 range

25.1 mean

5.6-80.0 range

115

15.9-501

113

8.1-386

Source: after White (1982)

29

Another approach to assessing primary productivity assumes a direct

relationship between phosphorus and chlorophyll concentration in the

water. However, lakes with high levels of phosphorus do not fit this

relationship (Stans, 1983). Studies by Vant and Pridmore (1981)

suggest that ATP (adenosine triphosphate) concentration may be a more

useful indicator of algal biomass. This is because when phytoplankton

numbers are high, chlorophyll increases faster than ATP in order to

compensate for increasing light limitation by self shading.

The phosphorus content of water is a function of inflow concentration,

residence time and loss to sediments. Estimates of phosphorus transfer

between Lake Taupo and the Waikato River suggest that 80-90% of phosphorus

inputs to the lake are retained in the lake (White and Downes, 1977).

Hoare (1982) has related the nutrient concentration in a lake to both

the nutrient concentration of the inflow and the 'retention coefficient'

of the lake. He has applied these concepts to lakes Rotorua, Tutira

and Alexandrina (Hoare, 1980; 1982; 1983). His results indicate that

mass balance phosphorus models can be very useful in predicting the

results of changes in nutrient loading or water retention time.

Potential sources of nutrients can be compared using simple calculations

before decisions on the treatment of eutrophication problems are made

(e.g. McColl, 1978a).

4. CONCLUSIONS AND POLICY IMPLICATIONS

The New Zealand agricultural industry is based to a considerable extent

on a ryegrass-clover system which requires substantial inputs of phosphate

to maintain productivity. Each year large quantities of phosphate

are applied to agricultural land in the form of superphosphate and

each year a proportion of this is lost via stock transfer, soil erosion

and water runoff.

Available data suggest that on farmland in many parts of New Zealand

losses through runoff and soil erosion can reach over 10% of the

maintenance fertiliser phosphorus additions.' The total value of phos­

phorus lost from agricultural land in runoff waters alone is conservatively

estimated as being $3-$6 million per year, while loss of phosphorus

30

through erosion of the topsoil by wind is likely to be equally

significant where farming practices expose dry silty and sandy soils

such as on the Canterbury plains and in Central Otago. Furthermore,

although overall estimates of phosphorus losses through animal

transfer are not known the few studies suggest that in hill country

at least an additional 5% per year can be lost in this way. Losses

of this kind are agriculturally very significant, costing in the order

of 15-25% of annual fertiliser expenditure. However, they may, with

appropriate management techniques, be partly avoidable.

Even when loss of phosphorus from agricultural land is less than 5%

of maintenance phosphorus additions, a level insignificant to farmers,

the effects of phosphorus on water quality of the receiving rivers

and lakes may be considerable. Water is the common property of the

people of New Zealand and hence adverse effects on water quality result

in losses that are shared by the community as a whole.

4.1 Future risks

Increased use of fertilizer in the future also implies increased losses

to the farmer via stock transfer, wind erosion and water runoff if

improved fertilizer management is not practised. If a 'do-nothing'

policy is adopted, the water quality of many New Zealand waterways

will continue to decrease from increased phosphorus loading.

Currently unaffected water bodies could also change in the future.

For lakes in agriculturally developed catchments, the expense and

effort of lowering trophic levels may be too great for society to bear.

However, lakes and their catchments can be managed to maintain the

present water quality so that they may continue to be used for recreation

and/or as a wildlife habitat •. Prevention is almost always less costly

than rehabilitation. It is therefore important to identify the potential

lake management problems in terms of the aspirations of people and

the technical and economic feasibility of fulfilling these aspirations.

The future risks and management needs are poorly understood in New

Zealand at the present time.

31

4.2 Management needs and approaches

If the surface runoff and wind erosion losses of fertiliser phosphate

could be controlled, using one or more of the methods put forward

in this paper, a significant reduction in the social costs associated

with water quality degradation could be achieved. For animal transfer

losses however, alternative grazing management approaches at present

are limited to significantly reducing overall fertiliser phosphorus

waste (inefficient use). Successful and economic fertiliser phosphorus

conservation methods would appear to involve increased diversification

rather than extremely intensive management currently precluded by

very high costs. Management options based on economic criteria from

the viewpoint of the individual farmer include "targeting" fertiliser

applications to those areas of hill country where stock transfer of

phosphorus is relatively less severe; silvipastoralism, which has

the added advantages, if well planned, of reducing runoff and erosion

and enhancing soil nutrient levels; and diversification e.g. running

goats or deer with, or rather than, sheep. However, while the techniques

for control of phosphorus losses listed in the second section of this

paper are all viable options few, if any, have been tested and

evaluated technically or economically in New Zealand and their relative

usefulness is unkown. This applies equally to the catchment develop­

ment phase (preventioro and the developed catchment phase (rehabilitation).

The economic benefit to the individual farmer of controlling losses

may not by themselves be sufficient to encourage farm phosphorus

conservation practices. But farm management may be the most

efficient way to avoid the social costs associated with reduced water

quality. Clearly, then, there is the important question of how to

encourage individuals to behave in such a way that contributes to

the larger interests of society. To achieve this end, the main

task will be to demonstrate to the farmer that wider land and water

management objectives are important and that various management options

are available to achieve them.

32

The environmental costs of farming are not generallyappreciated, particularly in the case of riparianwetlands. But it is certain that a positive, con­structive response would eventuate from the farmingcommunity, given encouragement and advice fromlocal authorities, water boards, and agriculturaladvisors. Regional and local government and theMinistry of Agriculture and Fisheries have especiallyimportant responsibilities in this respect.

Williams and Brickell, 1983.

Studies of the options from the viewpoints of efficient nutrient

control and agricultural management need to be undertaken by research

organisations so that authorities will have a basis on which to base

their recommendations to farmers.

33

ACKNOWLEDGEMENTS

The authors are very grateful to Dr R.H.S. McColl, T.J. Malthus

and Dr S.F. Mitchell for their criticism of the manuscript. We

also thank those who attended the phosphate workshop in February,

1984 and staff at the Centre for Resource Management for their

helpful discussions on phosphorus losses.

34

REFERENCES

Baker, J.L., Campbell, K.L., Johnson, H.P. and Hanway, J.J. 1975.

Nitrate, phosphorus and sulphate in subsurface drainage water.

Journal of Environmental Quality 4: 406-12.

Berg, N.A. 1980. Control of phosphorus from agricultural land in the

Great Lakes Basin. In: Phosphorus Management Strategies for

Lakes. Loehr, R.C., Martin, C.S. and Rast, W. (Eds.). Ann

Arbor Science, Michigan. 459-485.

Bolton, E.F., Aylesworth, J.W. and Hare, F.R. 1970. Nutrient losses

through tile drains under three cropping systems and two

fertility levels on Brookston clay soil. Canadian Journal of

Soil Science 50: 275-279.

Bostrom, B., Jansson, M. and Forsberg, C. 1982. Phosphorus release

from lake sediments. Archiv fur Hydrobiologie Beihefte

Ergebnisse der Limnologie 18: 5-59.

Bristow, J.M. and Whitcombe, M. 1971. The role of roots in the nutrition

of aquatic vascular plants. American Journal of Botany 58:

8-13.

Brown, J.M.A. and Dromgoole, F.I. 1977. The ecophysiology of

Lagarosiphon in Rotorua Lakes. Proceedings of the 30th New

Zealand Weed and Pest Conference 1977. 130-134.

Burwell, R.E., Timmons, D.R. and Holt, R.F. 1975. Nutrient transport

in surface runoff as influenced by soil cover and seasonal

periods. Proceedings of the Soil Science Society of America

39: 523-528.

Cooke, J.G. 1979. The effect of land use on water quality in New

Zealand. In: Proceedings of a seminar on land use in relation

to water quantity and quality, Nelson, November 7-8, 1979.

Nelson Catchment Board and Regional Water Board. 40-60.

1981. Pollution from our pastures. Soil and Water 17(5): 13-15.

35

Denny, P. 1972. Sites of nutrient absorption in aquatic macrophytes.

Journal of Ecology 60: 819-829.

Fish, G.R. 1969. Lakes: the value of recent research to measure

eutrophication and to indicate possible causes. Journal of Hydrology

(New Zealand) 8: 77-85.

and Andrew, I.A. 1980. Nitrogen and phosphorus in the

sediments of Lake Rotorua. New Zealand Journal of Marine and

Freshwater Research 14: 121-128.

Gillingham, A.G. 1980. Phosphorus uptake and return in grazed, steep

hill pastures. New Zealand Journal of Agricultural Research

23: 313-21.

and During, C. 1973. Pasture production and transfer

of fertility within a long-established hill pasture. New

Zealand Journal of Experimental Agriculture 1: 227-32.

Goltermann, H.L. 1973. Natural phosphate sources in relation to

phosphate budgets: a contribution to the understanding of

eutrophication. Progress in Water Technology 2: 3-17.

Haumann, D. and Waite, T.D. 1978. The kinetics of phosphate removal

in small alkaline lakes by natural and artificial processes.

Water, Air and Soil Pollution 19: 291-313.

Hilder, E.J. 1966. Rate of turnover of elements in soils: the effects

of stocking rate. Wool Technology Sheep Breeding, C.S.I.R.O.

13(2): 11-16.

Hoare, R.A. 1980. The sensitivity of phosphorus and nitrogen loads,

of Lake Rotorua, New Zealand. Progress in Water Technology 12:

897-904.

1982. Lake nutrient load calculations: A management tool.-------Soil and Water 18(3): 14-17.

1983. Review of water and soil conservation research.-------1981. Water and Soil Miscellaneous Publication No.53: 47-48.

36

Howard-Williams, C. 1981. Studies on the ability of a Potamogeton

pectinatus community to remove dissolved nitrogen and

phosphorus compounds from lake water. Journal of Applied Ecology

18: 619-637.

Hughes, J.G., McClatchy, D. and Hayward J.A. 1971. Beef cattle on

tussock country. Lincoln papers in Resource Management No.1.

Lincoln College. 270p.

Klopatek, J.M. 1978. Nutrient dynamics of freshwater riverine marshes

and the role of emergent macrophytes. In: Freshwater Wetlands:

ecological processes and management potential. Good, R.E., Whigham,

D.P., Simpson, R.L. (Eds.) Academic Press, New York. 195-216.

~Lambert, M.G., Devantier, B.P., Nes, P. and Penny, P.E. 1985. Losses

of nitrogen, phosphorus and sediment in runoff from hill country

under different fertilizer and grazing management regimes.

New Zealand Journal of Agricultural Research 28: 371-379.

Lee, E.S., Powell, H.K.J., Taylor, M.C. and Tsao, N.Y. 1979. Concentratiol

of nutrients in runoff from agley podzol soil in South Island,

New Zealand. New Zealand Journal of Science 22: 35-8.

McColl, R.H.S., 1972. Chemistry and trophic status of seven New Zealand

lakes. New Zealand Journal of Marine and Freshwater Research

6: 399-447.

1977. Chemistry of sediments in relation to trophic

condition of eight Rotorua lakes. New Zealand Journal of Marine

and Freshwater Research 11: 509-523.

1978a. Lake Tutira: the use of phosphorus loadings

in a management study. New Zealand Journal of Marine and

Freshwater Research 12: 251-6.

1978b. Chemical runoff from pasture: the influence

of fertiliser and riparian zones. New Zealand Journal of Marine

and Freshwater Research 12: 371-380.

37

MCColl, R.H.S. 1982. Water quality in agricultural areas - the prospects

for management of diffuse pollution sources. In: Water in New

Zealand's future. Proceedings of the 4th National Water

Conference 1982. Auckland. 153-160.

1983. Farm runoff - the mechanisms and moderating

factors. Soil and Water 19(3): 31-35.

and Gibson, A.R. 1979. Downslope movement of nutrients

in hill pasture, Taita, New Zealand. II. Effects of season,

sheep grazing and fertilizer. New Zealand Journal of Agricultural

Research 22: 151-161.

and Hughes, H.R. (Comp.) 1981. The effects of land

use on water quality - a review. Water and Soil Miscellaneous

Publication No. 23: 59p.

and Syers, J.K. 1981. The effects of agriculture.

Grassland and cropland. In: The effects of land use on water

quality - a review. McColl, R.H.S. and Hughes, H.R. (Comp.).

Water and Soil Miscellaneous Publication No. 23: 21-23.

_______________ ; White, E. and Gibson, A.R. 1977. Phosphorus and

nitrate runoff in hill pasture and forest catchments, Taita,

New Zealand. New Zealand Journal of Marine and Freshwater

Research 11: 729-744.

Malthus, T.J. 1985. Long term eutrophication of Lake Mahinerangi.

Ph.D. thesis in Zoology, University of Otago.

Minshall, N., Nichols, M.S. and Witzel, S.A. 1969. Plant nutrients

in base flow of streams in south western Wisconsin. Water

Resources Research 5: 706-713.

Mitchell, S.F. 1971. Phytoplankton productivity in Tomahawk Lagoon,

Lake Waipori, and Lake Mahinerangi. Fisheries Research Bulletin

No.3 (New Zealand Marine Department, Fisheries Divisionk 87p.

38

Mitchell, S.F. 1975. Phosphate, nitrate, chloride in a eutrophic

coastal lake in New Zealand. New Zealand Journal of Marine and

Freshwater Research 9: 193-198.

and Galland, A.N. 1981. Phytoplankton photosynthesis,

eutrophication and vertical migration of dinoflagellates in

a New Zealand reservoir. Verhandlungen Internationale Vereinigung

fUr Theoretische und Angewandte Limnologie 21: 1017-1020.

N.Z. Meat and Wool Boards' Economic Service, 1984. Supplement to

the sheep and beef farm survey 1981-82. Publication No. 1893,

Wellington.

of the sheep and beef industry 1984-85.

Wellington.

1985. Annual review

Publication No. 1918,

O'Connor, K.F. 1981. Comments on Dr Floate's paper on grazing

effect by large herbivores. In: Terrestrial nitrogen cycles.

Clark, F.E. and Rosswall, T. (eds.). Ecology Bulletin

(Stockholm) 33: 707-714.

O.E.C.D. 1981. Environmental policies in New Zealand: a review.

Government printer, Wellington.

Painter, D.J. 1978. The pastoral scene. Soil erosion rates on New

Zealand farm land. Proceedings of the Conference on Erosion

Assessment and Control. New Zealand Association of Soil

Conservators 1978, 25-42.

Patterson, K.J. and Brown, J.M.A. 1979. Growth and elemental composition

of the aquatic macrophyte, Lagarosiphon major, in response to

water and substrate nutrients. Progress in Water Technology

11: 231-246.

39

Pittams, R.J. 1979. Riparian strip studies - a review. Hamilton

Science Centre unpublished internal report.

Prentki, R.T., Gustafson, T.D., and Adams, M.S. 1978. Nutrient

movements in lakeshore marshes. In: Freshwater wetlands: ecological

processes and management potential. Good, R.E., Whigham, D.F.,

Simpson, R.L. (Eds.). Academic Press, New York. 169-194.

Richardson, C.J. 1985. Mechanisms controlling phosphorus retention

capacity in freshwater wetlands. Science 228: 1424-6.

__________________ , Tilton, D.L., Kadlec, J.A., Chamie, J.P.M. and

Wentz, W.A. 1978. Nutrient dynamics of northern wetland ecosystems.

In: Freshwater wetlands: ecological processes and management

potential. Good, R.E., Whigham, D.F., Simpson, R.L. (Eds.)

Academic Press, New York. 217-241.

Richmond, C.J. 1978. Lake Rotorua macrophytes: standing crop and

nutrient content in late February 1978. Unpublished report to

Conservator of Wildlife, Rotorua .

~RYden,J.C. and Syers, J.K. 1973. Evaluation of the potential of

soils for the phosphorus enrichment of streams. Proceedings

of the Pollution Research Conference, Wairakei, New Zealand,

20-21 June, 1973. 481-493

and Harris, R.F. 1973. Phosphorus in

runoff and streams. Advance in Agronomy 25: 1-45.

Saunders, W.H.H., Koller, B.G., Gillingham, A.G. and Hill, L.J. 1981.

Getting better value from fertiliser from pasture. Ruakura

Farmers Conference 33: 39-44.

_~Schuman~ G.E., Spomer, R.G. and Piest, R.F. 1973. Phosphorus loss:;r",

from four agricultural watersheds on Missouri Valley loess.

Proceedings of the Soil Science Society of America 37: 424-427.

40

Scott, G. 1985. Phosphorus cycling in New Zealand pastoral systems:

A conceptual framework for improved understanding and management

of phosphorus resources. Centre for Resource Management

Discussion Paper. Lincoln College. (In press).

Skovlin, J.M. 1984. Impacts of grazing on wetlands and riparian

habitat: a review of our knowledge. In: Developing strategies

for rangeland management. National Research Council/National

Academy of Sciences. Westview Press Inc. Colorado, U.S.A. 1000-1103.

~ Sharpley, A.N. and Syers, J.K. 1976a. Phosphorus transport in surface

runoff as influenced by fertiliser and grazing cattle. New Zealand

Journal of Science 19: 277-282.

1976b. Potential role of earthworm

casts for the phosphorus enrichment of runoff waters. Soil

Biology and Biochemistry 8: 341-346.

1979a. Loss of nitrogen and phosphorus

in tile drainage as influenced by urea application and grazing

animals. New Zealand Journal of Agricultural Research 22: 127-131.

1979b. Effect of aerial topdressing

with superphosphate on the loss of phosphate from a pasture

catchment. New Zealand Journal of Agricultural Research 22: 273-277.

~~ 1979c. Phosphorus inputs into a stream

draining an agricultural watershed. II. Amounts contributed

and relative significance of runoff types. Water, Air and Soil

Pollution 11: 417-428.

and Gregg, P.E.H. 1978. Transport

in surface runoff of phosphorus derived from dicalcium phosphate

and superphosphate. New Zealand Journal of Science 21: 307-310.

Sloey, W.E., Spangler, F.L. and Fetter, C.W. Jr. 1978. Management

of freshwater wetlands for nutrient assimilation. In: Freshwater

Wetlands: ecological processes and management potential. Good, R.E.,

Whingham, D.F., Simpson, R.L. (Eds.) Academic Press, New York. 321-340

41

Spence, D.H.N. 1982. The zonation of plants in freshwater lakes.

Advances in Ecological Research 12: 37-125.

Stans, J.C. 1983. The contribution of predictive methods to the

scientific approach in EIA. Delft Hydraulics Laboratory,

Publication No. 308. 26p.

Stockdill, S.M.J. and Cossens, G.G. 1984. Soil earthworm introduction

methods and effects on production. AGLINK FPP 211, Ministry of Agricuj

and Fisheries, New Zealand. 2p.

Stoltenberg, N.L. and White, J.L. 1953. Selective loss of plant nutrients

by erosion. Proceedings of the Soil Science Society of America 17:

406-410.

Stout, V.M. 1981. Lake Alexandrina. Freshwater Catch No.ll: 12-13.

Syers, J.K. 1974. Effect of phosphate fertilisers on agriculture

and the environment. New Zealand Agricultural Science 8: 149-164.

Harris, R.F. and Armstrong, D.E. 1973. Phoshpate chemistry

in lake sediments. Journal of Environmental Quality 2: 1-14.

and Ryden, J.C. 1973. Losses of phosphorus from forest

watersheds. Fertilising Forests Workshop. University of

Waikato, Hamilton. 65-79.

Teirney, L. 1980. Tutira, a lake worth restoring. Soil and Water 16(1):

10-13.

Thorrol~ B.S.,O'Connor, K.F., White, J.G.H. 1985. Management influences

on sheep behaviour, dung distribution and soil phosphate.

Proceedings of the New Zealand Grasslands Association 46:

127-134.

Turner, M.A., Tillman, R.W. and Syers, J.K. 1979. Losses of phosphorus

and nitrogen forms in subsurface drainage water as influenced

by phosphate fertiliser, grazing animals, and irrigation

treatments. Progress in Water Technology 11: 287-294.

42

Twinch, A.J. and Breen, C.M. 1978. Enrichment studies using isolation

columns. II. The effects of phosphorus enrichment. Aquatic

Botany 4: 161-168.

Young, R.A., Huntrods, R. and Anderson, W. 1980. Effectiveness of

vegetated buffer strips in controlling pollution from feedlot

runoff. Journal of Environmental Quality 9: 483-487.

van der Valk, A.G., Davis, C.B., Baker, J.L. and Beer, C.E. 1978.

Natural fresh water wetlands as nitrogen and phosphorus traps

for land runoff. In: Wetland Functions and values: the state

of our understanding. Greeson, P.E., Clark, J.R., Clark, J.E.

(Eds.) American Water Resources Association, Minneapolis. 457-476.

Vant, W.N. 1985. Lake Rotorua: nutrients regenerated from anoxic

bottom sediments. Paper presented to New Zealand Limnological

Society Conference, Auckland, 1985.

and Pridmore, R.D. 1981. Nutrients and phytoplankton in

four Waikato lakes. In: Waters of the Waikato. Campi C

and Ridell, G. (Eds.). Proceedings of Seminar, University

of Waikato, 1981. 115-133.

Vincent, W.F. and Downes, M.T. 1980. Variation in nutrient removal

from a stream by watercress (Nasturtium officinale R.Br.)

Aquatic Botany 9: 221-235.

Vollenweider, R.A. and Dillon, P.J. 1974. The application of the

phosphorus loading concept to eutrophication research.

National Research Council of Canada No.13690. 42p.

Von Oertzen, I.A.M.L. 1981. Use of different water plants in biological

filter systems for the treatment of agricultural wastewaters.

B.Ag.Sc. thesis. Rheinische Friedrich-Willhelms Universitat,

Bonn, West Germany.

43

White, E. 1982. Eutrophication in New Zealand Lakes. In: Water in

New Zealand's Future. Proceedings of the 4th National Water

Conference, 1982, Auckland. 129-136:

1983. Lake eutrophication in New Zealand - a comparison

with other countries of the Organisation for Economic Co-operation

and Development. New Zealand Journal of Marine and Freshwater

Research 17: 437-444.

__________ , Don, B.J., Downs, M.T., Kemp, K.J., Mackenzie, A.L. and

Payne, B.W. 1978. Sediments of Lake Rotorua as sources and sinks

for plant nutrients. New Zealand Journal of Marine and Freshwater

Research 12: 121-130.

__________ , Downes, M.T. 1977. Preliminary assessment of nutrient

loads on Lake Taupo, New Zealand. New Zealand Journal of Marine

and Freshwater Research 11: 341-356.

Williams, P. and Brickell, D. 1983. Guideline Riparian zone management.

Upper Waitemata Harbour Catchment Study. Auckland Regional Authority. I!

Witzel, S.A., Minshall, N, Nichols, M.S. and Wilke, J. 1969. Surface

runoff and nutrient losses of Fennimore watersheds. Transactions

of the American Society of Agricultural Engineering 12: 338-341.

Yates, M.E. 1971. Effects of cultural changes on Makara experimental

basin: Hydrological and agricultural production effects of

two levels of grazing on unimproved and improved small catchments.

Journal of Hydrology (New Zealand) 10: 59-84.