Murky Waters: multidimensional issues linkingmacronutrient sources & impacts in catchments

Helen Jarvie, Andrew Sharpley, Paul Withers

Exploring uncertainties & expectations.....

(1) Point sources are being addressed, so we need tofocus efforts on diffuse transfers to meet river waterquality & ecological targets

(2) Catchments are inherently leaky: P losses fromagriculture are a primary cause of nuisance algal growthin rivers

(3) Reducing P losses from agriculture will achievedesired improvements in river water quality & ecology

Beliefs driving catchment management policy:

A UK river water quality perspective

Use of background tracers: boron (detergents)as a tracer of sewage effluent

Across 54 major UK lowland rivers: Pshows a dominant sewage fingerprint

For P, dissolved reactive (DRP) concentrations during periods of ecologicalsensitivity pose the greatest risk for nuisance algal growth in lowland rivers

0 20 40 60 80 100

500

1000

1500

2000

120

Flow (m3s-1)

DR

P(µ

g-P

l-1)

Thames

…Coupled with flow dependence:

Jarvie et al, 2006. Science of theTotal Environment, 360, 246– 253

Dilution with flow – dominance ofpoint sources at low flows

.....BUT even in headwater agricultural catchments, point sources dominate TPloads most of the time (especially under baseflows, periods of ecological sensitivity)

Jarvie et al, 2010. Agriculture, Ecosystems and Environment 135, 238-252.

Yes, diffuse sources dominate annual TP loads......

Successes in P remediation: point source P controls

Ju

n-9

7

Se

p-9

7

De

c-9

7

Ap

r-98

Ju

l-98

Oct-9

8

Ja

n-9

9

Ma

y-99

Au

g-9

9

No

v-99

Ma

r-00

Ju

n-0

0

Se

p-0

0

De

c-0

0

Ap

r-01

Ju

l-01

Oct-0

1

Fe

b-0

2

Ma

y-02

Au

g-0

2

0

100

200

300

400

500

600

700

800

DR

P(μ

gl-

1) Upgrade to STW (P-stripping)

River Kennet

River reach mass balance studies (R. Lambourn):%

ch

an

ge

inD

RP

load

-100

-50

0

50

100

150

200

DRP release

DRP retention

May Aug Dec

2003

Feb June Oct

2004

Recovery phase~8 months

‘Legacy P’: responses to point source P remediation in rivers

P inputremoved

Short water & sediment residence times - rapid recovery

Jarvie et al., 2006. J Hydrology, 330, 101– 125

In-l

ake

Pco

ncen

trati

on

(µg

L-1

)

0

50

100

150

200

250

300

1990 1995 2000 2005 2010

2010

Point-source P inputsreduced 60%

Recovery phaseInternal recycling of ‘legacy’ P

(10-15 y)Waterquality

target met

Water quality target40 µg/L

Legacy P: longer residence times mean delayed WQ responses in lakes

Loch Leven, Scotland (Source: Linda May, CEH)

1992 2010

Farmyard

Spring House roofwater

Field drain

Road drain

• Multiple sources & forms of P

• ‘Landscape filtering’ of P along thewatershed-river continuum

• Storage & re-release of ‘legacy’ P

Longer legacy ‘lag’ effects after agricultural controls?

Evaluating P retention & release at the watershed scale

Extended End-member Mixinganalysis (E-EMMA)

Simple & versatile tool, reliessolely on routinely-measured Pconcentration and flow data.

Up to 50% retention of annualTotal P loads

Up to 80% retention of Total Ploads under ecological-criticallow flows (spring & summer)

Buffering along the watershed-river continuum regulatesdelivery of P to help reduceecological impactsdownstream

J. Environmental Quality, 40, 492-504

BMP successes: Conservation tillage

No-till reduced erosion from wheat fields (2 ha) by 95%

Convertedto no-till

1980 1985 1990 1995

Conventionaltill wheat

6

4

2

0

-

-

Total P (mg L-1)

From: Sharpley and Smith, 1994 – El Reno, OK

Runoff

1980 1985 1990 1995

20

15

10

0

-

5

-

Total N (mg L-1)

Convertedto no-till

Conventionaltill wheat

-

Conservation tillageConservation tillage

Maumee River - Annual flow-weighted Total P (mg/L)

Dave Baker & Peter Richards, OH

50% decrease

1975 1985 1995 2005

0.8

0.6

0.4

0

0.2

Adoption of mulch and no-till soybeans, %Annual flow-weighted dissolved P, ppm

75% decrease

80

60

40

20

Trends in P – Maumee River

1975 1985 1995 2005

0.12

0.09

0.06

0

0.03

Process mechanisms: coupled macronutrient cyclesD

epth

(cm

)

SRP (µg-P/l) NH4 (mg/l) SO4 (mg/l)

Palmer-Felgate et al (2011), Sci. Tot. Environ., 409, 2222-2232

P thresholds for nuisance algal (periphyton) growth

Bowes et al (2007) Canadian J. Fisheries and Aquatic Sciences 64 (2): 227-238

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 50 100 150 200 250 300 350 400 450

SRP conc (μg L-1)

Ch

loro

ph

yllc

on

c(s

tan

dar

dis

ed

)

P-stripped Control P-addition

River Thames

100 µg-P L-1

Even murkier - ecological responses to P remediation

DR

P(μ

gl-

1)

800

100 µg-P L-1

0

100

200

300

400

500

600

700River Kennet

DR

P(μ

gl-

1)

1997: Healthy chalk stream ecosystem; SRP c. 600 µg L-1

1999-2005: proliferation ofnuisance algae; SRP c. 80 µg L-1

Loss of invertebrate grazers?Increased stocking of brown trout

Ecosystem response?

Murky waters....

Aquatic ecologistsSoil scientists

Water quality scientists

Policy makers

Biogeochemists

• Resource & aquatic habitat management• Wider range of physico-chemical controls on river ecology• Coupled macronutrient cycles & ‘self-cleansing’ capacity of rivers

Current focus on diffuse-source P controls & nutrient criteria alonemay not necessarily achieve the desired ecological & water qualityoutcomes within short policy-relevant timescales….

But opportunties for more integrated approaches....

Source: Colin Neal, CEH & Ian Bateman, UEASource: Colin Neal, CEH & Ian Bateman, UEA

Which do you prefer?

Source: Colin Neal, CEH & Ian Bateman, UEASource: Colin Neal, CEH & Ian Bateman, UEA

1980 1985 19951990

0.5

1.0

0

Infiltration increased 33%

Convertedto no-till

Runoff - Dissolved P (mg L-1)

Conventionaltill wheat

Conservation compromises

1980 1985 19951990

Nitrate (mg L-1)

30

20

10

0

Convertedto no-till

Conventionaltill wheat

Leached -