The Science Behind the Phosphorus Risk Index

Wes Jarrell, Professor and HeadNatural Resources and

Environmental Sciences, UIUC March 2, 2005

-from Robertson et al.

EPA proposed Total P criteria for Upper Midwest water bodies

Forms of P -

% of total WaterMenomonee R, Armstrong

Dissolved (PO4)Sorbed

Insoluble/inorganicCa, Fe, Al -P

Organic

0.04%

15%

60%

25%

30%

3%

27%

40%

TP =0.138 mg P/L TP= 500 mg P/L

Soil

?

Readily bioavailable

Seasonably bioavailable

?

e.g., Bray

Instantly bioavailable: dissolved PO4 and desorbable PO4

Seasonally bioavailable P: P in organic or inorganic particles that is released over a growing season

Load vs concentration

Sources of water to water body

• Runoff - main source of P load

• Direct precipitation inputs

• Baseflow - seepage from groundwater:

• Wisconsin: 10 - 50 ppb total P

• Higher concentrations in some areas

• Point sources - discharges from municipal-industrial sources

FieldConcentrated flow

Particles – Entrainment

Settling - enrichment

Receiving water

PrecipitationP Transport – Land to Water

Nonpoint sources

Delivery zone

Dissolved Dissolution Sorption/Desorption

Rusle2Bray vs totalBray vs solubleSoluble P in fert.

Sedimentdelivery ratioBuffer effectiveness

Settling - enrichment

SUMMARY

 

Phosphorus that could cause problems in water occurs both in particles (PP) and dissolved in solution (SP) when it reaches surface water.

We consider both particles and dissolved P.

Land/water/animal manager

Researcher

APEX - field/farm

P index/ Rusle2

PI-EZ

PALM - field/watershed SWAT -

watershed

Phosphorus Transport - Land to Water

LEVEL OF DETAIL USER

Desirable Characteristics of P index

- Accurately rank fields in order of their risk of supplying P to a water body

- Based on best available science, easily modified to reflect improvements

- Easy to use, interpret, and apply

- Helps user understand factors affecting P movement to water

- Direct user to improved management practices that effectively and economically lower the risk

- Should be applied over the whole farm

- Provide maximum flexibility to farmer, while decreasing P loading.

Total Risk Index for Phosphorus (PI):

 

PI = PP + SP + LP

PI = Total P index

PP = Particulate P

SP = Soluble P

LP = Leached P

General location of soil samples Total soil P Total P in clay Enrichment overmg P/kg mg P/kg whole soil

Forested silty sw WI 427 1627 3.8Prairie silty sw WI 731 1502 2.1Forested silty se WI 391 852 2.2Prairie silty se WI 562 2136 3.8Central Sands 149 1656 11.1W WI sandstone 232 2007 8.7Red sandy and loamy, e WI 298 654 2.2Acid sandy loamy, N and E WI 304 1405 4.6Forested silty N WI, ign/metam, N and E WI 212 248 1.2Forested sandy glacial drift 217 807 3.7

Averages 352 1289 4

Boerth and Helmke, 1997

Total P in soils and clays, native soils, Wisconsin (Boerth and Helmke, 1997)

1,000 2,000 3,000 4,000 5,0000

Average P concentration in particulates, mg P/kg

Wisconsin soils (Boerth and Helmke)

Wisconsin soil clays (Boerth and Helmke)

Runoff plot sediment (Bundy and Andraski)

Small Wisconsin streams (Baum, WDNR)

Wisconsin streams (Corsi et al., USGS)

Living algal cell or crop plant leaf

Soil organic matter

Scenescent leaf

Modeling land use effects (Panuska, others, WDNR)

Manure

To 11,000+

PP: Depends on (1) erosion, (2) fraction of eroded particles delivered to stream, and (3) P concentration in the soil particles

 

Calculation:

Particulate P =

Rusle2 *

Sediment Delivery Ratio *

Enrichment Ratio *

BufferEffectiveness*

Soil particle P concentration (from Bray P1)

Pheasant Branch: TP vs total P, Elena Bennett

y = 3.153x + 341.98

R2 = 0.6721

0

200

400

600

800

1000

1200

1400

1600

0 50 100 150 200 250 300 350

Bray P1, mg/kg

To

tal P

, m

g/k

g

Also:Meyer, Lyne, Avila, Barak, UW Madison, Plano silt loam: Total P = 2.5 (Bray P1) + 875

“It appears that most of the sediment generated by a particulate erosion

event is usually deposited in small or headwater tributaries.”

- Glymph and Storey, 1967

Sediment Delivery Ratio

S = SY0 e – T (D)1/2

Where S = sediment yield at the down stream channel outlet,SY0 = sediment yield at the upstream end of the channel,

= (Beta) decay constant or routing coefficient,T = Travel time through the section in hours,D = the particle diameter in millimeters.

From John Panuska – J.R. Williams original model

Soluble P: Depends on amount of runoff, P concentration in the soil, and soluble P concentrations in P-containing amendments/fertilizers

Total soluble P =

SP from soil P

+ SP from unincorporated nutrients on unfrozen soil

+ SP from unincorporated nutrients on frozen soil

( + SP release from crop residues?)

 

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00.0

P concentration in solution, mg/L

Upper limit for oligotrophic lake (total)

Lower limit for eutrophic lake (total)

Lower limit for maximum crop growth in soil (soluble)

Upper limit for oligotrophic stream (total)

Soil solution [P] at 50 ppm Bray P-1

Lower target for wastewater discharge(total )

Manure: 10 - 30 mg P/L(100:1 dilution)

a. Soluble P load in runoff:

For no fertilizer or incorporated fertilizer:

Soil solution equilibrium [P] (from Bray P1) *

Annual runoff volume (from Rusle2)

Bray P1 vs soil solution P

y = 30.1x + 17.4

R2 = 0.901

0

50

100

150

200

250

0.00 2.00 4.00 6.00 8.00

Soil solution [P], mg P/L

Bra

y P

1, m

g P

/L

Bundy and Andraski

b. Soluble P load in runoff –

For surface-applied nutrients without incorporation:

Soluble P in manure/fertilizer (lb/A) /

average days between runoff-generating events

b. Soluble P load in runoff – cont’d

Soluble P in manure

Phosphorus forms in dairy manures (Good and Cooperband)

Manure source Total P Total TDP/TP(TP) dissolved P (TDP)mg/kg dm mg/kg dm

Dilution ratio100:1 1000:1 100:1 1000:1

Liquid 9500 1280 5900 0.14 0.62

Sand bedding 2200 1425 1530 0.64 0.69

Organic bedding 6500 1980 2718 0.33 0.44

4 “Rainfall 2000 Runoff Events 2000

N-S Chisel Plow

Runoff Events 2000

Contour Ridge TillRunoff event frequency

Arlington, Wisconsin

PALM output, Norman et al.

c. Soluble P in runoff from frozen soil –

For snow-melt events with nutrients spread on frozen soil:

Soluble P in manure (lb/acre)*

Slope percentage(squared)/200

**NOT YET INCORPORATED INTO MODEL**

d. Soluble P load from crop residues

For loss from crop residue

SP = f(Soluble P in residue,

spring runoff volume, ??)

**NOT YET INCORPORATED INTO MODEL**

LP = P lost through leaching, especially to tile lines

LP = f(P concentration in soil solution, depth to tile, retention coefficient, recharge volume)

 

Farm Name:County WinnebagoField ID BD12Other ID (tract, etc)2001 crop cornAcres 7.8Bray P1 300Erosion 5Calc Slope 4Unincorporated P fertlizer, % P2O5 (18-46-0)Application rate per acre in poundsManure Analysis P205 (available) (3 or 8 default) 3Manure Analysis % Moisture 87Manure Rate on NON-FROZEN soil (tons) 0Manure Rate on NON-FROZEN soil (gal) 0Manure Rate on FROZEN soil (tons) 0Manure Rate on FROZEN soil (gallons) 0Distance to Concentrated Flow 0Crop Rotation c-c-c-o-h-h-hSoil Types HoB-OaB-OaC-BrBResidue mgt FPRR

Calculated PI 7.26Particulate PI 5.46Soluble PI 1.68Leaching PI 0.12Whole-farm index 2.75Whole-farm particulate PI 2.02Whole-farm soluble PI 0.61Basic farm dataNumber of milking cows 255Months of manure storage % of manure spread on frozen ground.

 

P index values for one field:Corn, 4% slope, 300 ppm Bray P1, 50% sol P in manure

Tillage M* M M NT** NT

Erosion 5 5 5 0.5 0.5Distance to concn. flow 1000 1000 0 1000 1000Manure, T/A 0 25 0 0 25

Total PI 5.8 8.2 7.3 2.2 4.6Particulate PI 4.0 4.0 5.5 0.4 0.4Soluble PI 1.7 4.1 1.7 1.7 4.1LP 0.1 0.1 0.1 0.1 0.1

*M = Moldboard plow**NT = No Till

Land Use Loadinglb/acre/year

Low Most likely High

Row crop agriculture 0.50 1.00 3.00Mixed agriculture 0.30 0.80 1.40Pasture/grass 0.10 0.30 0.50High density urban 1.00 1.30 2.00Medium density urban 0.30 0.50 0.80Rural residential 0.05 0.10 0.25Wetlands 0.10 0.10 0.10Forest 0.05 0.09 0.18

Export coefficients – ModelPanuska et al., WDNR

Preliminary Interpretation of risk associated

with PI values:

0 – 2 Low risk: low probability of being a problem except for very sensitive water bodies

2 – 6 Intermediate risk: important for water bodies sensitive to P inputs

6 - 10 High risk: Likely excessive in most watersheds

>10 Very high risk: Excessive for almost any water body

The Phosphorus Risk Index -Progressive Planning

1 2 3 4 5 6 7

Years

PhosphorusRiskIndex

Case 1 - degradation

Case 2 - balance

Case 3 - restoration

The framework of the PI is in place.

Now it needs

(1)ADAPTATION to specific regional conditions in Wisconsin;

(2)EVALUATION at a variety of scales, to see if it truly measures what we intend it to measure;

(3)USABILITY for planners and plan implementers;

(4)COMPARISON with other states

Summary

-The PI should be part of a Systems Approach to Phosphorus Management in Wisconsin Agriculture

- It provides a framework into which we can incorporate the best existing science and extend it to users

-The PI should complement and be consistent with other models

- Gaps in our understanding of the system are identified through applying the PI

- Adaptation, Evaluation, Usability, and Comparability are needed to apply the PI more efficiently

Forms of Phosphorus in Water

and Their Bioavailability

Wes Jarrell, Senior Scientist

Discovery Farms Program

"A river and its plankton are a flowing soil and its crop,...." (p. 147).

Forbes, S. A. and R. E. Richardson. 1919. Some recent changes inIllinois River Biology. Bull. Ill. State Natural History Survey13:141-156.(Thanks to Erwin Van Nieuwenhuyse, CA)

Corollary:"A soil is benthic sediment in an intermittent stream supporting emergent vegetation.”

W.M. Jarrell, 2000, and likely someone else, circa 1940.

“It has been suggested that God must have been a limnologist or an oceanographer …”

Harris, Phytoplankton Ecology, 1986

Regulations will be based on Total P in water; we have no easy ways of determining bioavailable P in the particle fraction

Total P (mg P/L) =Particulate P (PP) + SRP

Loads

Total P: All P in particles and all dissolved P entering water body

Bioavailable P: Phosphorus that is, or can rapidly become, the PO4 (ortho-P) form; from 10 to 90+% of total P

Follow the colloids!

Colloid:

- Particle less than 2 m in diameter

(clays, organic matter)

- Settles out of water very slowly

- High surface to volume ratio

- Often high concentrations of nutrients

Total P (mg P/L) = Particulate P (PP) + SRP

Total P in water is comprised of

(1) Particulate P, which does NOT pass through the 0.45 m filter,and

(2) P that passes through a 0.45 m filter,also called “dissolved P”, “soluble reactive P” (SRP), “ortho-P”, “soluble P”

(reminder: colloids at usually <2m)

Actual analyses

Total P: strong acid dissolution of entire water sample, determine ortho-P in digest.

SRP: direct determination of water passing through a 0.45 m filter

SRP - Immediately bioavailable

- Where did it come from?

- How can we control it?

Unitsppb: g/L (for water), g /kg (particles, soil); orppm: mg/L (for water) or mg/kg (particles, soils)

Minimum solution concentrations from which plants can extract Pin flowing solution:

Algae: 0.3 - 0.6 - 1 g P/L (ppb), 0.001 ppm

Rye: 3 g P/L (ppb), 0.003 ppm

Oats: 7 g P/L (ppb), 0.007 ppm

Critical P concentrations and trophic state in water

Periphyton streams (low flow)

Trophic state Total P

mg P/L

Eutrophic 0.020

Bioavailability

BrayP-1 vs soil solution P

y = 30.126x + 17.371

R2 = 0.9007

0

50

100

150

200

250

0.00 2.00 4.00 6.00 8.00

Soil solution [P], mg P/L

Bra

y P

-1,

mg

P/L

Bundy and Andraski

PARTICULATE PHOSPHORUS

- Particulate P (PP) in water is organics, aluminum, iron, and calcium phosphates - How much PP is bioavailable in a given situation? - Where did it come from? - How can we control it?

To estimate average P concentration in suspended particles in water:

(1) Calculate Particulate [P]: PP(mg P/L) = TP - SRP

(2) Divide PP concentration by TSS concentration:(Total suspended solids (TSS) in water: mg solids/L)

P concentration insuspended solids

PP (mg P/L)TSS (mg solids/L)= = mg P

mg solids

Ecoregion Total suspended solids Total P TP/TSSMedian Median (assume TP = 70% particulate))

Tons/acre Lb/acre mg P/kg particulate

Northern Lakes and Forest 0.11Driftless Area 0.33 1.37 1319SE WI Till: Rural 0.05 0.44 2797SE WI Till: Urban 0.20 0.50 795State Summary: Rural 0.17 1.02 1907State Summary: All 0.19 0.78 1305

SOURCE: USGS Fact Sheet FS-195-97: Corsi, Gracczyk, Owens, Bannerman

P concentrations in water particulates

Site Total P (lbs)/Sediment (tons) Total P, mg/kg Estimated Particulate Pconcentrations (assume 30% DRP)

Beaver Creek 7 3,500 2,450Otter Creek 7 3,500 2,450Brewery Creek 5 2,500 1,750Garfoot Creek 6 3,000 2,100Eagle Creek 2 1,000 700Joos Valley 2 1,000 700

Ken Baun, 1995

P concentrations in water particulates

Land cover Total P Total solids P in particulatekg/ha kg/ha (assuming particulate P

is of 70% total P)High density urban 1.34 595 1576Low density urban 0.34 350 680Golf courses 0.22 280 550Ag - cranberry bog 0.69 140 3450Grassland 0.34 70 3400Forest 0.10 35 2000Open water 0.34 35 6800Wetland 0.10 18 3889Barren 1.46 630 1622Shrubland 0.11 10 7700

Watershed water quality monitoring,Panuska and others, WDNR

P concentrations in water particulates

- How do TP concentrations in soil compare with PP in water?

- Does this explain why we can’t simply add up RUSLE across a watershed and get good quality water?

- How does this help us compare different P pollution sources - P credit trading?

Why doesn’t it work to add up all the RUSLE values for a watershed to get total lost?

- Sediment delivery ratio

- Enrichment ratio

- Unknown relationships between Bray P1and total P

(1) Sediment delivery ratio (SDR)

As water and sediment moves from land to water, larger particles drop faster

e.g., in perfectly still water, to drop 20 cm requiressand: 2 minutessilt: about 2 hoursclay (colloids): 8 hours to > 1 year

In turbulent water, water energy keeps larger particles in suspension, especially colloids

(2) Enrichment ratio

As heavy particles drop out, lighter particles (especially colloids) stay in suspension:

“bed load”: heavy sandy particles“wash”: particles that are light, colloidal -

organics, clays - that move with water

NOTE: Colloids and most organic matter have much higher P concentrations than do sands and some silts.

(3) Different soils have different TP concentrations

- Very little known about total P in Wisconsin soil particles

- For same soil type, should be a correlation between Bray P-1 and total P in the soil

- Reactive clay-sized particles have a much higher [P] than silts and sands

y = 223943x-1.7733

R2 = 0.6487

0

2

4

6

8

10

12

14

16

18

0 500 1000 1500 2000

[P] in solids, mg/kg

To

tal

susp

end

ed s

oli

ds,

g

/kg

Runoff plots: TSS in runoff vs [P] insolids (Bundy and Andraski)

Stratification from Robertson, Saad, and Wieben, 2001

103 ug/L37 ug/L

125 ug/L286 ug/LMean Conc.

Movement from land to water

What determines how much P is mobilized?• Breaking away from aggregates:

• strength of aggregate bonds (organic matter);• energy of water

• Movement• energy of water• size of particles;

• density of particles

Movement from land to water

Load vs concentration

Soil has a wide range of particle sizes and density

Most of bioavailable P is associated with

(1) small particles, colloids; and

(2) light particles, organic matter (about 1/3 the density of clay)

Bulksoil

Dissolved

Particulate

Dissolved

Particulate

Algae

Settledheavy/largeFiltered

Smaller,

lightercolloids

Sorbed

“enrichment”

Movement from landto water - BMP function

Desorption

Uptake

Settledheavy/large

Issues

Load vs concentration

- If soluble levels are increasing in the environment, will our current BMPs work?

- If clays and small organics are the primary runoff particulates, where will BMPs work?

Issues (cont’d)Load vs concentration

- If soluble levels are increasing, which of our current BMPs will work?

- Lower soil soluble P: P management, chemical treatments?

- Lower manure soluble P: feed and chemical management

- Increase infiltration/lower runoff: organic matter, plant management

Issues (cont’d)

Load vs concentration

- If clays and small organics are the primary runoff particulates, which BMPs work best?

- Increase infiltration

- Improve aggregation, wet strength

- Increase residue

BETTER USE OF ORGANIC MATERIALS

AND PERENNIAL PLANTS

Issues (cont’d)

Load vs concentration

- QUESTION:

Will buffers work to decrease the problems with runoff?

- Yes, for physical effects: sands and silts, bottom deposition, abrasion

- Not for removing colloids: except by increased infiltration

Recommendations- Determine actual forms of particulate phase P in runoff, manures, water column, and bottom sediments

- Evaluate current BMPs based on their effects on colloidal particle transport and soluble P transport

- Develop new BMPs for manure, land management, and in-water management targeted at mitigating colloidal and soluble P impacts.

Land to water transport -

It IS a continuum!

Bioavailable P from solids vs total sediment loss

y = 0.0333x - 2.2293

R2 = 0.8437

-50

0

50

100

150

200

250

0 1000 2000 3000 4000 5000 6000

Total sediment loss, kg/ha

BAP - DRP, kg/ha s