Leaf-litter application to a sandy soil modifies phosphorus leaching over the wet season of southwestern Australia

Primary Research Paper

Leaf-litter application to a sandy soil modifies phosphorus leaching over the

wet season of southwestern Australia

Song Qiu*, Arthur McComb, Richard Bell & Jenny DavisMurdoch University, School of Environmental Science, Sout Street, 6150, Perth, WA, Australia(*Author for correspondence: E-mail: [email protected])

Received 04 June 2004; in revised form 16 January 2005; accepted 29 January 2005

Key words: phosphorus leaching, nutrient retention, P cycling, Eucalyptus litterfall, microbial activity, southwestern

Australia

Abstract

Nutrient leaching is a critical step in terrestrial litter turnover, and is potentially linked to nutrient cycling indownstream wetlands. Little is known about the leaching behaviour and P loading from terrestrial litterunder conditions of winter rainfall in southwestern Australia. In this study, leaf litter of flooded gum(Eucalyptus rudis Endl.), a species common in southwestern Australia, was exposed for leaching underwinter rainfall. Litter P was leached primarily during the first few weeks of the wet season, and the ‘firstflush’ generating a mean P load of 114.7 mg m)2 from the litter. Thereafter the P leached decreasedprogressively with the progress of the wet season. Overall, P leaching was correlated with rain intensity,with 84.1% of litter P leached over the wet season May to November. When litter was applied to a bare,sandy soil and then subjected to rain leaching, more P was leached during the ‘first flush’ compared with thedata from ‘litter only’ leaching, but a portion of P released from litter and soil appeared to be retainedthrough litter–soil interactions. Litter application to soil surface and via surficial burial reduced leachate Pby 25.2–29.5% and 28.6–38.6%, equivalent to a P retention of 75 and 81 mg P m)2, for surface application(10 cm soil) and surficial burial (5 cm soil), respectively. The P retention was attributed to increasedmicrobial immobilisation, supported by increased nutrient flux from litter.

Introduction

Plant litter from terrestrial and riparian vegetationserves as a nutrient and carbon source for down-stream waters via direct litterfall, wind, run-offand seepage from fringing wetlands (Brinson et al.,1980; McClain & Richey, 1996). This source mayaccount for over 70% of annual soil carbon flux(Raich & Schlesinger, 1992), and is well known asa carbon source for streams and riverine floodplains in Australia (Robertson et al., 1999).

The turnover of litter proceeds typically via twomajor steps, an initial rapid weight and nutrientloss primarily due to leaching, and later structuraldisintegration and decomposition primarily due tofungal and bacterial activity (Valiela et al., 1985;

Xiong & Nilsson, 1997; Eisenbeis et al., 1999).Leaching usually occurs rapidly when litterencounters flooding water (Willoughby, 1974;Taylor & Parkinson, 1988), and was reported torelease about 25% N and 50% P from the leaflitter (Parsons et al., 1990), or even depleted nu-trients except N during the first month (Furtado &Verghese, 1981). Litter decomposition andnutrient release are time-dependent and related tolitter types and quality (O’Connell & Menage,1982; Cornelissen, 1996; Cortez et al., 1996; Aerts& Decaluwe, 1997), influenced by decompositionenvironment conditions (e.g., temperature andmoisture regime, pH, oxygen) especially thoseassociated with microbial community development(Swift et al., 1979; O’Connell, 1994; Tate, 2000).

Hydrobiologia (2005) 545:33–43 � Springer 2005DOI 10.1007/s10750-005-1826-5

In wetland and floodplain environments, hydro-period (i.e. inundation frequency and duration) isan important factor governing ecological processes(Mitsch & Gosselink, 1993; Reddy et al., 1999),altering the soil environment and influencingmicrobial organisms to process organic matter(Tate, 1980; Neckles and Neill, 1994; Lockabyet al., 1996).

Compared with the considerable literature onspecies- and site-specific litter decomposition andassociated rate-controlling factors and theireffects, little is known about terrestrial litter as anutrient source for adjacent aquatic systems,especially under a mediterranean climate of sea-sonal rains. In southwestern Australia, >90% ofannual rainfall is concentrated in winter monthsfrom May to October, coinciding with mostnutrient loads being delivered to local waters(Donohue et al., 2001). On a regional scale, thetopsoils are severely leached, infertile, and typi-cally contain low P and organic matter (McAr-thur, 1991); freshwater wetlands are generally Plimited in relation to the algal growth, and the Psupply has been largely attributed to humanactivities such as agriculture and urbanizationduring the last few decades (Davis et al., 1993).Although litter P leaching occurs annually duringthe wet season, little is known about the movementof litter-sourced P or the ‘off-site’ potential of litternutrients to affect wetlands downstream. Wetlandsin the region are surface expressions of the shallowgroundwater (Townley et al., 1993), which meansthat litter leachate from the catchment may beconveyed to low-lying wetlands within a ground-water flow network.

Qiu et al. (2002) reported that litterfall in areasfringing wetlands leached an average of 30% totalP in 24 h. The data suggest that the amount of Pleached from litter via annual rainfall may equalthe P held in the water column of shallow wetlandson an area (catchment)-to-area (wetland) bases(Qiu et al., 2002). A difficult question has beenrelated to the movement of leachate P duringwinter rains on these low-gradient sandy land-scapes, which are expected to have a relatively highinfiltration. The critical issues include the leachingpatterns of litter P under the seasonal rains (themain driving force of the P movement), and theinteractions between leaching forces and the fac-tors modifying P leaching during the rainy season.

Concentrations of P leached from surface littermay be modified during infiltration and transportthrough soil layers via leachate-soil interactions.Understanding such processes is essential to assessP export potential from these sandy soils, not onlyfor from these wooded landscapes but also inagricultural systems where litter–soil–nutrientinteractions are expected to occur at a higher level.

This paper reports the results from a litterleaching experiment conducted in the field underwinter rainfall conditions of southwestern Aus-tralia to address such issues. Leaf litter fromEucalyptus rudis Endl. growing in a woodlandcatchment was exposed to winter rains, with orwithout contact with the bare soil from thecatchment. Leachate from the experimental unitswas monitored over the wet season. The majorfactors associated with litter–soil interactions,including the method of litter application (on soilsurface or burial) and the infiltration depth ofleachate in soils (5 or 10 cm) were examined inrelation to their potential for modifying net Pleaching. Implications of P retention via litter–soilinteractions during the wet season are discussed.

Materials and methods

Meteorological conditions

The region is under mediterranean climate, withhot dry summers and wet and mild winters, whichhave mean daily temperatures from 13–24 �C. Themean daily maximum temperature ranges from17 �C in July to 30 �C in January and February.The region receives strongly seasonal rainfall. Ofan average annual rainfall of about 800 mm,about 90% falls in four winter months (from Juneto September); this is followed by a period ofalmost no rain from November until April of thefollowing year. Potential evaporation is high dur-ing ‘dry months’, and low in ‘wet months’.

Litter and soil collection

Litter and soil samples were collected from thesoutheast catchment of Thomsons Lake, nearPerth, Western Australia. The lake is a sub-coastalfreshwater wetland, one of a chain of wetlands ininterdunal depressions between the Bassendean

34

and Spearwood Sand Dune systems of the SwanCoastal Plain, southwestern Australia.

Litterfall from flooded gum (Eucalyptus rudisEndl.), a species common in the region, was col-lected in March 2002 using litter traps before theonset of the wet season. Leaf material was air-dried and stored in paper bags at room tempera-ture. A portion of plant material was dried andground to about 50 lm, and stored in polystyrenevials in a desiccator at room temperature fornutrient analysis.

Previous study on this catchment suggests adifferent level of soil organic content and microbialactivity between the surface layer (0–5 cm) and thelayers below (Qiu et al., 2003). Surface soil (0–5 cm, termed ‘Soil A’) and sub-surface soil (5–15 cm; termed ‘Soil B’) were therefore collectedfrom the upland catchment, in an area withouttrees and with little surface vegetation. Soils weremanually separated according to depth using a flatshovel. Soil samples were well mixed, a portion air-dried and its water content recorded.

Litter leaching in absence of soil

The leaching test was conducted in duplicate di-rectly in the field under prevailing wet seasonconditions. Briefly, 15 g dw of litter were weighedinto nylon mesh bags (mesh size ca. 0.5 cm, toprevent wind blowing), and spread evenly on a net‘shelf’ 15 cm above the bottom the polyethyleneleaching tank (16.2 cm · 20.6 cm · 18 cm height),as such that the litter was maintained in a non-

flooded condition. The amount of litter applica-tion (surface coverage) was based on the estima-tion of annual litter production (0.5 kg leaf litterm)2 yr)1) determined for this catchment using lit-ter traps in the year before this study.

Care was taken to minimize potential inputs ofP from other sources. The experimental units wereplaced on a metal framed ‘table’ more than 1 mabove the ground, and was covered by nylon net toprevent large incoming objects. The P content inrainwater was generally below <2l g l)1 duringthe experimental period according to our randomobservations.

Litter leaching in the presence of soil

Soils collected at field moisture were transferredinto 20 leaching tanks (in duplicates) and gentlycompacted (by shaking and bumping) to a depthof 5 or 10 cm (equivalent to 1.65 and 3.30 kg dw,respectively), each then receiving 15 g (dw) ofEucalyptus leaf litter, applied according theexperimental design (Table 1).

All leaching tanks were exposed to seasonalrainfall, at the experimental field site at MurdochUniversity, 8 km north west of Thomsons Lake, toparallel with the weather data recorded at Mur-doch weather station. The leaching experimentcommenced 19 April 2002, before the commence-ment of winter rain; the last leachate sample wascollected on the 1st of November, at the end of thewet season. Due to high evaporation (572 mmrainfall cf. 459 mm evaporation during the study

Table 1. Experimental design for litter–soil leaching under field rainfall. Leaf litter (Eucalyptus rudis) was applied to soil surface or

1 cm below surface (surficial burial) on a rate of 0.5 kg leaf litter m)2

Leaching tanks Soil origin Reconstructed depth (cm) Litter application

Litter with soil A 10 On soil surface

B 10 On soil surface

A 5 On soil surface

B 5 On soil surface

Litter with soil A 5 Surficial burial

B 5 Surficial burial

Soil control A 10 Without litter

B 10 Without litter

A 5 Without litter

B 5 Without litter

35

period), the leaching tanks were usually in a ‘noleachate’ condition unless there is a period ofconcentrated rainfall. Over the study period theleaching tanks, including ‘litter only’, ‘litter withsoil’, and ‘soil control’ were drained six times, on9-May, 2-June, 11-June, 8-July, 28-August,1-November, following significant periods of rain,to avoid prolonged flooding. For ‘litter only’leaching tanks the leachate was directly sampledusing a pipette following mixing; for the ‘litter withsoil’ leaching tanks and their soil controls, leachatewas drained by syphoning from a ‘bore’ (a poly-ethylene pipe of 4 cm diameter) installed in thesoil, screened with nylon mesh (ca. 0.2 mm) at thebottom. The bores were capped with a water-tightscrew-on cap between sampling intervals. The‘drain volume’ from each tank was recorded eachtime, homogenized and sampled for leachate P, pHand colour (E440).

Leachate collected at each sampling wasimmediately stored at 4 �C for chemical analysis.pH, anion-exchange-membrane extractable P(AEM-P) and blue colour absorption at 440 nm(E440) were measured within 48 h. Total P wasthen measured as soon as practicable, generallywithin a week of sampling.

Chemical and statistical analysis

Soil water content and pH were measured the dayafter sampling, using the method of Rayment andHigginson (1992). Organic carbon was determinedby H2SO4 and dichromate oxidation (Rayment &Higginson, 1992). AEM-P was directly extractedfrom leachate, by shaking 20 ml of leachate withan AEM strip (2 · 2 cm) in 20 ml distilled water(16 h). Retained P was eluted in 0.1 N H2SO4, andmeasured by molybdenum-blue spectrophotome-try (Kouno et al., 1995; Qiu et al., 2002). Total P inleachate and soils was measured by molybdenum-blue spectrophotometry after perchloric aciddigestion. Plant elemental analysis was carried outon ground samples using inductively coupledplasma spectrometry (Zarcinas et al., 1987) afterHNO3 digestion. Reference plant materials wereused for analytical quality control.

P leached from ‘litter only’ experimental unitsand soil controls units (soil only) were mathe-matically aggregated, and the effect of soil-litterinteractions was estimated using the difference of P

leaching between sum of ‘litter only’ and ‘soil only’(which excludes soil–litter interactions) and ‘litterapplied to soil’ (which includes the interactions).The sum of P leached from ‘litter only’ and ‘soilonly’ (termed as separate leaching) can be ex-pressed using the following formula:

Separating leachingðmgm�2Þ

¼ 1

A

Xn

i¼1PliVli þ

1

A

Xn

i¼1PsiVsi

where n is the number of the drains or the periodssampled; Pli is total P concentration of ‘litter only’leachate during the period (i) (mg l)1); Vli is thevolume of ‘litter only’ leachate drained during theperiod (i) (L); A is the surface area of leaching tank(m2); Psi is total P concentration of the ‘soil only’leachate during period (i) (mg l)1); Vsi is the vol-ume of ‘soil only’ leachate drained during thecorresponding period (L).

Total seasonally leached-P (May to November)was estimated by integration of P leached duringeach of the six periods. ‘Rain intensity’(mm day)1) was calculated as rainfall (mm)divided by the length (days) of the period. Statis-tical analysis was conducted to compare timevariable leaching results between soils, and be-tween application methods, using the statisticalpackage (e.g. ANOVA two-factor with replica-tion) on Microsoft Excel 2000. To explore thefactors associated with P leaching, correlationanalysis was conducted among leachate properties(pH, colour, total P), and among leachate prop-erties and rainfall parameters (volume, duration,and intensity) using regression and correlationmodules on Microsoft Excel 2000.

Results

Litter leaching without soil

Litter P was primarily leached out within the firstfour weeks under non-inundated conditions, from6 May to 2 June (Fig. 1). Thereafter there waslittle change in leachate P concentration. Overall,84.1% of total P in the litter was leached outduring the wet season under field rainfall condi-tions. P leaching was positively correlated withrain intensity (n = 6; R2 = 0.67, *p<0.05).

36

Litter applied to soil

Soils used in the leaching experiment were greyishloamy sand, with little organic carbon and total P(Table 2). The soils contained a little inorganic N(<2 mg kg)1 NO3–N and NH4–N). Soils A and Bhad similar pH and AEM-extractable P. Both werevery dry when collected from the field.

When litter was applied to soils (surface appli-cation or surficial burial) and subjected to rainfallleaching, most P (average 55.9%) was leachedduring the ‘first flush’ (38 mm rain fell within aweek) of the wet season. The averaged leachate Pconcentration during the first flush was 5.1 mg l)1.The ‘first flush’ from litter and soil was followed bya relatively smaller leaching peak between June andJuly. Compared with the aggregated leaching re-sults from litter (in absence of soil) and soil (inabsence of litter), litter applied to 10 cm soil(‘combined’) leached less P from either Soil A orSoil B in early period (Fig. 2). More P was leachedduring a later phase (8 July) in ‘combined’ leaching,but the overall result for the wet season is that litterapplied to the soil surface leached less P at 10 cmdepth, both in Soil A and Soil B (Table 3).

With surface application of litter, Soil Awasmoreeffective in reducing leaching than Soil B at 10 cm

depth – significantly less total P leached in Soil Athan in Soil B over the wet season (Fig. 3a; ANOVA:between soils F = 7.10, *p=0.03; between timesF = 58.2, ***p < 0.001; interaction F = 1.53,p = 0.28). The total of P leached over the wet season(sum of periods) was less in Soil A than in Soil B(Table 3). The large p value of the interaction sug-gests that the effects of soil types (A and B) on Pleaching did not vary with time over the wet season.

Soil depth (reconstituted soil) appears to be afactor affecting P leaching results; less P leached at10 cm depth than at 5 cm, either in Soil A(Fig. 3b; ANOVA: between depth F = 185.19,***p < 0.001; between time F = 63.59, ***p <0.001; interaction F = 40.08, ***p < 0.001) or inSoil B (Table 3; ANOVA: between depthF = 9.03, *p=0.02; between time F = 21.43,***p < 0.001; interaction F = 5.67, *p = 0.02).The response of soil depth to time varied signifi-cantly in relation to P leaching.

The method of litter application appearedcritical for P retention: Surficial burial reducedoverall P leaching (sum for the season) at 5 cm soildepth of either Soil A or B, compared with theaggregated results where litter and soil were lea-ched separately (Fig. 3c and Table 3. ANOVAbetween methods F = 67.28, ***p<0.001 for Soil

Table 2. Properties of the soils used in the leaching experiment

Original layer

collected (cm)

Water content % pH Organic C % AEM-P

(mg kg)1)

MB-P

(mg kg)1)

Total P

(mg kg)1)

Soil A 0–5 1.08 6.0 0.70 1.37 9.13 50

Soil B 5–15 0.94 6.2 0.41 1.08 3.00 26

MB-P – microbial biomass-P.

0

20

40

60

80

100

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov 1-Dec

Time

Lit

ter

P r

emai

nin

g (%

)

0

2

4

6

8

10

12

14

Rai

nfa

ll (m

m)

P% remaining

Rainfall

Figure 1. Percentage of total P in Eucalyptus leaf litter leached under the wet conditions. Dot line leads to the initial state where there

was no leachate produced. Error bar indicates ± standard error (SE) of the measurements.

37

A; F = 6.82, *p=0.03 for Soil B). The response ofapplication method varied with time in relation toP leaching.

Leached total P and colour (Fig. 4) were corre-latedwith rain intensity (R2=0.67)0.92, *p<0.05),irrespective of soil type, depth and litter applicationmethod; Overall, leachate pH appeared to increasewith increased time (days) of the individual periodsexposed to leaching, and decrease with increasedrain intensity. In 5 cm soils leachate total P wasinversely correlated with pH, irrespective of soiltype, depth and litter application method(R2=0.90)0.92, *p<0.05). This relation, however,was not significant in leachate at 10 cm in Soil A, orsoil B (R2=0.15)0.42) that have demonstrated Pretention, nor in soil controls.

Discussion

P leached from litter

The amount of nutrients released through litterturnover may vary with vegetation type, land use,

Soil A

0

50

100

150

200

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov

Time

P le

ach

edm

g m

–2

Soil B

0

50

100

150

200

250

300

Time

P le

ach

ed m

g m

–2

Combined

Separate

Combined

Separate

(a)

(b)

Figure 2. Total P leached (mean ± SE) when litter applied to soil (‘combined’); or litter leached in absence of soil and soil leached in

absence of litter and then results aggregated (‘separated’). (a) soil A 10 cm depth, combined vs. separte. Litter only data in dotted line.

(b) Soil B 10 cm depth, combined vs. separate.

Table 3. Total P leached (sum of the season, mean ± SE) from

‘litter applied to soils’ over wet season, differing between soil

types, depth, and application method; and from ‘litter and soil

(separate leaching)’, where litter leached in absence of soil and

soil leached in absence of litter and the results aggregated

Soil type Leaching

method

Total P leached

over the wet

season (mg m)2)

Litter applied to soils

Soil A (10 cm) Surface 176 ± 5

Soil B (10 cm) Surface 218 ± 2

Soil A (5 cm) Surface 287 ± 5

Soil B (5 cm) Surface 189 ± 2

Soil A (5 cm) Burial 175 + 23

Soil B (5 cm) Burial 146 ± 3

Litter and soil (separate leaching)

Soil A (10 cm) Separate 235 ± 11

Soil B (10 cm) Separate 308 ± 11

Soil A (5 cm) Separate 245 ± 10

Soil B (5 cm) Separate 238 ± 10

In each case Eucalyptus leaf litter was applied on a rate of

0.5 kg leaflitter m)2.

38

litter quality, and climate conditions (Molineroet al., 1996; Aerts, 1997; Xiong & Nilsson, 1997).In this study, the between-tank difference in litterquality and ‘site conditions’ was considered negli-gible, because a single litter type was used and allleaching tanks were exposed in a same environ-ment in relation to temperature and evaporationetc. Each tank received the same amount of rain-fall as the primary forcing for nutrient release over

the wet season. There were only a few light rainshowers from February to April. Dry weather stilldominated the catchment and soil moisture waslow (average 6.3%). It was thus not possible togenerate infiltration of litter leachate until May toJune, when the major rain commenced and 38 mmrain fell within the first week, generating a P fluxfrom litter (the first flush). The rapid leaching anddepletion of leachable P by early rains suggests

Effect of reconstructed depth

0

50

100

150

200

250

300

Time

P le

ach

ed m

gm

–2

10 cm

5 cm

Effect of application method

0

50

100

150

200

250

300

Time

P le

ach

ed m

gm

–2

surface application

burial

Effect of soil layer

0

30

60

90

120

150

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov

Time

P le

ach

ed m

gm

–2

0

2

4

6

8

10

12

14

Rai

nfa

ll m

mSoil A (0-5 cm)

Soil B (5-15 cm)

Rainfall

no leachate

(a)

(b)

(c)

Figure 3. Effect of soil layers (depth of soil sampling), reconstructed depth and litter application method on total P leached (mg m)2):

(a) Soil A vs. Soil B, both 10 cm reconstructed depth and surface application. (b) 5 cm vs. 10 cm reconstructed depth. Both soil A with

surface application. (c) Surface application vs. bury. Both soil A 5 cm reconstructed depth. Error bar indicates SE of the measure-

ments.

39

that much of the litter-P on the catchment, aftermonths of hot and dry weather, was readily solu-ble or mobilizable by rainwater. This ‘first flush’resulted in about 67.1% of the total P leachedfrom the flooded gum leaflitter, and was equivalentto a P load of 114.7 mg P m)2 to surface soils.Overall, the total P concentration in litter leachateduring the first flush (4.1 mg l)1 in absence of soil)was comparable with the P concentrations in ur-ban runoff (typically 0.1–3 mg l)1, O’Loughlin etal., 1992). The volume of the leachate was howeverrelatively small since the rain volume producingthe first flush was only about 7% of the total rainvolume of the wet season. Nevertheless such a Pload on soil surface would conceivably have directinfluence on litter and soil microbial activities.

Using litter traps we recorded approximately0.3 kg m)2 yr)1 of leaflitter in this woodlandcatchment, and the average P content of the leaflitter was 0.5 g kg)1. Assuming an average 60% oflitter total P leached in field conditions over thewet season (our unpublished data on four commonlitter species of the catchment), the leachable Pfrom average litter was estimated to be0.91 kg P ha)1 on the woodland wetland catch-ment, which equals to 9.3–29.4% of the annualfertilizer input in the farmed catchment of the

region (3.1–9.8 kg P ha)1 yr)1, Birch (1982). Sucha P component cannot be overlooked in P cyclingand mass balance in these P-deficient soils. Previ-ous studies in the region indicated that thepotential export from diffuse agricultural sourceswas 0.11–1.67 kg P ha)1 yr)1, and those exportedfrom the farmed catchments was 0.12–0.89 kg P ha)1 yr)1 (Gillingham & Thorrold,2000; Birch, 1982). Such P export from farmlandhas caused a significant eutrophication in estuariesof southwest Australia in the past (McComb,1995). Our estimated P load from litter, as dis-cussed above, was comparable in size to the Pexport data from farming sources, and may thusconstitute a significant P source to water bodies ifthe leached P were exported. The critical issue ishow much litter-P can be exported, and how muchlitter-P would be retained in soil via leachate-soilinteractions during the wet season.

Litter–soil interactions

We found both Soil A and Soil B of 10 cmreconstructed depth able to detain a portion ofleached P. The reduction of leachate P (as com-pared with aggregated results of ‘separate leach-ing’) via surface application was 25.2–29.5%,

0.00

0.40

0.80

1.20

1.60

2.00

0.00

0.40

0.80

1.20

1.60

2.00

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov 1-Dec

1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov 1-Dec

Time

Lea

chat

e co

lou

r (E

440)

Lea

chat

e co

lou

r (E

440)

0

4

8

12

16

20

Rai

nfa

ll m

mR

ain

fall

mm

10 cm soil

5 cm soil

Soil only

Rainfall

Time

0

5

10

15

20Surface

Burial

Rainfall

(a)

(b)

Figure 4. Rainfall and the colour leached from ‘litter applied to soils’: (a) surface application leached at different depths and soil

control; (b) surface application and surficial burial leached at 5 cm depth.

40

being dependent on soil quality and the depth ofsoil profile interacting with litter and leachate.Surficial burial appeared more effective than sur-face application in P retention and reduced 28.6–38.6% of total P in the leachate. On average of thetwo soils layers (0–5 and 5–15 cm), the resulted Pretention over the wet season was 75 and81 mg P m)2, for surface application (10 cm soil)and surficial burial (5 cm soil), respectively. Itappears that much of the P (ca. 1/4–1/3) releasedfrom litter and soil would be retained in topsoilsvia soil–litter interactions under the wet seasonconditions of south-western Australia.

The higher P retention via surficial burial(compared with surface application) appears to bea result from increased microbial immobilization.Previous study on litter–soil interactions in thiscatchment suggested that heterotrophic microbialbiomass may capture most ‘leachable P’ (AEM-P)during early rains, resulted in a three-fold increasein microbial biomass-P following the commence-ment of wet season, 4.8–43.9 mg kg)1 of P insurface soils retained by soil microbial biomass,and only about 5% of ‘leachable P’ remaining insoils as compared with that held in microbialbiomass (Qiu et al., 2003). Aggangan et al. (1999),who compared the effects of ‘surface’ and ‘mixed’application of Eucalyptus leaf litter on N leachingin a native forest and a pasture soil of the region,and demonstrated ‘no leaching’ over 29 weekswhen litter was mixed with soil, and incubated andleached in laboratory with simulated ‘rains’; thereduction in N leaching was attributed to micro-bial immobilization. Elsewhere, application ofwheat straw, wheat stubble and sheep manure(5.22 g kg)1 soil) to soil induced rapid N immo-bilization, and no net N-mineralization wasdetected until day 52 (Corbeels et al., 1999).Changes in the substrate conditions especiallymoisture, carbon and nutrients appears to be theunderlying cause of microbial immobilization.

Compared with above soils, Soils A and B inthis study are very infertile, especially low in car-bon and nutrient contents (Table 2). The site haspreviously been used as ‘background’ site forcharacterizing litter-related P leaching and het-erotrophic microbial activity along a transect fromupland to wetland, and thus supported little het-erotrophic microbial activity. The correspondingmicrobial activity (CO2 efflux) from this site was

an order of magnitude lower than at other sitesalong the transect (Qiu et al., 2003). The C/P ratioof Eucalyptus leaflitter from this catchment wasabout 416 (Qiu et al., 2002), and the C/P ratio ofsoils used in this study was about 140. The highC/P ratios of litter and soil relative to soil micro-bial cell (C/P ratio typically about 49; source ofIOWA State University) means that there wouldbe sufficient C available to soil microbial cells andthus a high likelihood for microbial biomass toimmobilize P, rather than to mineralize organicmatter for their C assimilation needs (Mullen,1998).

The potential contribution from physico-chemical adsorption to the observed P retentionwas considered to be small in our study. The studycatchment, lies in a depression between two seriesof coastal sand-dunes of Pleistocene origin, ischaracterized by a general lack of Fe materials andpoor P adsorption in the upper horizon of soilsbecause of prolonged weathering and leaching(McArthur, 1991). Recent studies reaffirmed thatthere is little P adsorption in surface soil in thearea, associated with low Fe-oxides content (Heet al., 1998; Allen et al., 2000; Qiu et al., 2004). Inthis study the significant P leaching (high P con-centrations in leachate) in early wet season (May)was followed by a gradually depletion of leachableP towards the end of the wet season. The P con-centrations of the leachate were virtually negligiblein November, which implies that P adsorption tosoil particles (assuming P adsorption occurs), as aresult of higher equilibrium concentration duringthe early stage (May), would have been graduallydesorbed and depleted as P equilibrium concen-tration decreased to near zero (September–November). Soil water held after periodic rainsmay retain a pool of leached P. P held in soil waterduring an early stage, however, would also bedepleted via rain leaching towards the end of thewet season, and thus have little effect on theoverall results of P leaching.

Conclusion

Exposure of flooded gum leaf litter to the winterrains of southwestern Australia released 84.1% ofits total P over the wet season. Significant P wasreleased during the first flush, accounted for 67%of litter P content. When litter was applied to the

41

surface of a sandy soil from the catchment andleachate was monitored at 10 cm soil depth of theleaching tanks, less P was found in leachate com-pared with that of the 5 cm soil depth, or theestimated results excluding soil-litter interactions.Litter application through surficial burial resultedin even less P in the leachate. Overall, soil–litterinteractions reduced P leaching by 25.2–38.6%,equivalent to P retention of 75 and 81 mg P m)2

on surface application (10 cm soil) and surficialburial (5 cm soil), respectively. The P retentionwas attributed to increased microbial immobiliza-tion nourished by a nutrient flux from litterleachate.

Acknowledgements

This work forms a part of on-going study supportedby ARC Large Grant A00105241. The Departmentof Conservation andLandManagement ofWesternAustralia issued permits for field studies.

References

Aerts, R. & H. Decaluwe, 1997. Nutritional and plant-mediated

controls on leaf litter decomposition of carex species. Ecol-

ogy 78: 244–260.

Aerts, R., 1997. Climate, leaf litter chemistry and leaf decom-

position in terrestrial ecosystems: a triangle relationship.

Oikos 79: 439–449.

Aggangan, R. T., A. M. O’Connell, J. F. McGrath & B. Dell,

1999. The effects of Eucalyptus globulus Labill. leaf litter on

C and N mineralization in soils from pasture and native

forest. Soil Biology and Biochemistry 31: 1481–1487.

Allen, D., G. S. Mann & I. R. Wilson, 2000. Managing Soil

Nutrients in Residential Areas of the Swan Coastal Plain. A

Preliminary Report. Chemistry Centre, Western Australia,

Perth, Australia 31.

Birch, P. B., 1982. Phosphorus export from coastal plain

drainage into the Peel-Harvey estuarine system of Western

Australia. Australian Journal of Marine and Freshwater

Research 33: 23–32.

Brinson, M. M., H. D. Bradshaw, R. N. Holmes & J. B. Elkins

Jr., 1980. Litterfall, stemflow, and throughfall nutrient fluxes

in an alluvial swamp forest. Ecology 61: 827–835.

Corbeels, M., G. Hofman & O. Van Cleemput, 1999. Simula-

tion of net N immobilization and mineralisation in substrate-

amended soils by the NCSOIL computer model. Biology and

Fertility of Soils 28: 422–430.

Cornelissen, J. H. C., 1996. An experimental comparison of leaf

decomposition rates in a wide range of temperate plant

species and types. Journal of Ecology 84: 573–582.

Cortez, J., J. M. Demard, P. Bottner & L. J. Monrozier, 1996.

Decomposition of Mediterranean leaf litters – a microcosm

experiment investigating relationships between decomposi-

tion rates and litter quality. Soil Biology and Biochemistry

28: 443–452.

Davis, J. A., R. S. Rosich, J. S. Bradley, J. A. Growns, L. G.

Schmidt & F. Cheal, 1993. Wetland Classification on the

Basis of Water Quality and Invertebrate Community Data.

Wetlands of the Swan Coastal Plain, Vol. 6. Water

Authority of Western Australia, Perth 242.

Donohue, R., W. A. Davidson, N. E. Peters, S. Nelson & B.

Jakowyna, 2001. Trends in total phosphours and total

nitrogen concentrations of tributaries to the Swan-Canning

Estuary, 1987–1998. Hydrological Processes 15: 2411–2434.

Eisenbeis, G., R. Lenz & T. Heiber, 1999. Organic residue

decomposition: The minicontainer-system – A multifunc-

tional tool in decomposition studies. Environmental Science

and Pollution Research 6: 220–224.

Furtado, J. I. & S. Verghese, 1981. Nutrient turnover in a

freshwater inundated forested swamp, the Rasel Bera,

Malaysia. Verhandlugen – Internationale Vereinigung fur

Theoretische und Angewandte Limnologie 21: 1200–1206.

Gillingham, A. G. & B. S. Thorrold, 2000. A review of New

Zealand research measuring phosphorus in runoff from

pasture. Journal of Environmental Quality 29: 88–96.

He, J. Z., R. J. Gilkes & G. M. Dimmock, 1998. Mineralogical

properties of sandy podzols on the Swan Coastal Plain,

South-West Australia, and effects of drying on their phos-

phate sorption characteristics. Australian Journal of Soil

Research 36: 395–409.

Kouno, K., Y. Tuchiya & T. Ando, 1995. Measurement of soil

microbial biomass phosphorus by an anion exchange

membrane method. Soil Biology and Biochemistry 27:

1353–1357.

Lockaby, B. G., A. L. Murphy & G. L. Somers, 1996.

Hydroperiod influences on nutrient dynamics in decompos-

ing litter of a floodplain forest. Soil Science. Society of

American Journal 60: 1267–1272.

McArthur, W. M., 1991. Reference Soils of South-western

Australia. Department of Agriculture, Western Australia,

Perth, Australia.

McClain, M. E. & J. E. Richey, 1996. Regional-scale linkages of

terrestrial and lotic ecosystems in the Amazon basin. A

conceptual model for organic matter. Archiv Fur Hydrobi-

ologie 113:111–125.

McComb, A. J., 1995. Eutrophic Shallow Estuaries and La-

goons. CRC Press, USA.

Mitsch, W. J. & J. G. Gosselink, 1993. Wetlands, 2nd edn. Van

Nostrand Reinhold, New York.

Molinero, J., J. Pozo & E. Gonzalez, 1996. Litter breakdown in

streams of the Aguera catchment-inflence of dissolved

nutrients and land use. Freshwater Biology 36: 745–756.

Mullen, M. D., 1998. Transformation of other elements. In

Sylvia, D. M. J. J. Fuhrmann, P. G. Hartel, & D. A.

Zuberer (eds), Principles and Applications of Soil Microbi-

ology. Prentice-Hall, New Jersey: 369–383.

Neckles, H. A. & C. Neill, 1994. Hydrologic control of litter

decomposition in seasonally flooded prairie marshes. Hyd-

robiologia 286: 155–165.

42

O’Connell, A. M. & P. M. A. Menage, 1982. Litterfall and

nutrient cycling in karri (Eucalyptus diversicolor F. Muell).

forest in relation to stand age. Australian Journal of Ecology

7: 49–62.

O’Connell, A. M., 1994. Decomposition of nutrient content of

litter in a fertilized eucalyptus forest. Biology and Fertility of

Soils 17: 159–166.

O’Loughlin, E. M., W. J. Young, J. D. Molloy, 1992. Urban

Stormwater: Impacts on the Environment. CSIRO Divisions

of Water Resources Consultancy Report 92/29.

Parsons, W. F. J., B. R. Yaylor & D. Parkinson, 1990.

Decomposition of aspen (Populus tremuloides) leaf litter

modified by leaching. Canadian Journal of Forrest Research

20: 943–951.

Qiu, S. & A. J. McComb, 2004. Groundwater induced accu-

mulation of iron oxides and phosphorus retention in severely

leached soils. Marine and Freshwater Research 55: 213–222.

Qiu, S., A. J. McComb &R.W. Bell, 2002. Phosphorus leaching

from litterfall in wetland catchments of the Swan Coastal

Plain, southwestern Australia. Hydrobiologia 472: 95–105.

Qiu, S., A. J. McComb, R. W. Bell & J. A. Davis, 2003.

Nutrient response to soil and litter metabolic activity in a

transect across a seasonal wetland. Marine and Freshwater

Research 54: 243–252.

Raich, J. W. & W. H. Schlesinger, 1992. The global carbon

dioxide flux in soil respiration and its relationship to vege-

tation and climate. Tellus 44B: 81–99.

Rayment, G. E. & F. R. Higginson, 1992. Australian Labora-

tory Handbook of Soil and Water Chemical Methods. Ink-

ata Press, Melbourne.

Reddy, K. R., R. H. Kadlec, E. Flaig & P. M. Gale, 1999.

Phosphorus retention in streams and wetlands: a review.

Critical Reviews in Environmental Science and Technology

29: 83–146.

Robertson, A. I., S. E. Bunn, P. I. Boon & K. F. Walker, 1999.

Sources, sinks and transformations of organic carbon in

Australian floodplain rivers. Marine and Freshwater

Research 50: 813–829.

Swift, M. J., O. W. Heal & J. M. Anderson, 1979. Decompo-

sition in Terrestrial Ecosystems. Studies in Ecology, Vol. 5.

University of California Press, Berkeley, CA.

Tate, R. L. III, 2000. Soil Microbiology. John Wiley & Sons

Inc., New York.

Tate, R. L. III, 1980. Microbial oxidation of organic matter in

histosols. Advances in Microbiology and Ecology 4: 169–

201.

Taylor, B. R. & D. Parkinson, 1988. Annual differences in

quality of leaf litter of aspen (Populus tremuloides Michx).

affecting rates of decomposition. Canadian Journal of Bot-

any 66: 1940–1947.

Townley, L. R., J. V. Turner, A. D. Barr, M. G. Trefry, K. D.

Wright, V. Gailitis, G. J. Harris & C. D. Johnston, 1993.

Wetlands of the Swan Coastal Plain. Interaction Between

Lakes, Wetlands and Aquifers Water Authority of Western

Australia, Perth, Australia 115.

Valiela, I., J. M. Teal, S. D. Allen, E. R. D. Van Goehringer &

D. Volkmann, 1985. Decomposition in salt marsh ecosys-

tems: the phases and major factors affecting disappearance

of above-ground organic matter. Journal of Experimental

Marine Biology and Ecology 89: 29–54.

Willoughby, L. G., 1974. Decomposition of litter in Fresh-

water. In Dickinson, C. H. & G. J. F. Pugh (eds), Biology of

Plant Litter Decomposition, Vol. 2. Academic Press, New

York: 659–682.

Xiong, S. J. & C. Nilsson, 1997. Dynamics of leaf litter accu-

mulation and it effects on riparian vegetation – a review.

Botanical Review 63: 240–264.

Zarcinas, B. A., B. Cartwright & L. R. Spouncer, 1987.

Nitric acid digestion and multi-element analysis of plant

material by induced coupled plasma spectrometry. Com-

munications in Soil Science and Plant Analysis 18: 131–

146.

43