Effect of land use and moisture on phosphorus ... - CSIRO Publishing

CSIRO PUBLISHING

www.publish.csiro.au/journals/mfr

Marine and Freshwater Research, 2009, 60, 619–625

Effect of land use and moisture on phosphorus forms in upland stream beds in South Otago, New Zealand Richard W. McDowell AgResearch, Invermay Agricultural Centre, Private Bag 50034, Mosgiel 9053, New Zealand. Email: [email protected]

Abstract. Land use can influence stream sediment composition and water quality, whereas moisture status affects sediment phosphorus (P) bioavailability to algae. Declining upland surface-water quality in South Otago, New Zealand, may reflect land-use changes from sheep- to dairy-farming. I sampled sediment (0–20 cm) from streams draining 12 dairyand 12 sheep-farmed catchments in spring (wet) and autumn (dry). 31 P nuclear magnetic resonance (NMR) spectroscopy and the EDTA-fractionation scheme were used to determine different P forms and infer P bioavailability. Significantly more P was present in the sediment of streams draining dairy- than sheep-farmed catchments. Total P did not differ with the moisture regime; however, changes occurred in the following P fractions: acid-soluble organic P, NaOH-P, CaCO3 ≈P, Fe(OOH)≈P and residual organic P. Extraction for 31 P NMR analysis removed 78–85% of sediment total P and isolated five P classes. More bioavailable P such as orthophosphate (23–40% of P extracted) and diesters (2–6% of P extracted) was present in dry than in wet sediments, and in sediments draining dairy streams than in those from sheep-farmed catchments. This indicates substantial reserves of bioavailable P in sediment from these catchments, especially from dairy-farmed catchments, sustaining in-stream P concentrations for many years even without additional P input from land. Additional keywords: depth, diesters, fractionation, monoesters, 31 P NMR.

Introduction Stream-bed sediments can act as sinks or sources of phosphorus (P) and play a central role in the production of nuisance bacteria and algae in aquatic systems. Annual wetting and drying cycles have a profound effect on the P dynamics of abiotic and biotic sediments. For example, De Groot and Van Wijck (1993) found that when anoxic wetland sediments were drained, the oxidation of previously reduced mineral phases (e.g. iron (Fe) oxides) increased P sorption and the mineralisation of organic P. However, Howell et al. (1998) noted that if large quantities of Fe sulfides were present then oxidation may decrease pH and solubilise some mineral phases. With sustained desiccation, several studies have shown that sediment affinity for P is decreased, owing to mineral aging of oxyhydroxides (e.g. Sah et al. 1989; Qiu and McComb 1994). Desiccation can also kill up to threequarters of the resident microbes (Qiu and McComb 1995), and on rewetting, release P previously held within microbes and biota, causing an increase in productivity that may ultimately lead to anoxia. Alternatively, if incomplete desiccation or continual wetting and drying cycles occur, then it is likely that a microbial community tolerant of moisture stress would develop (Fierer et al. 2003; Humphries and Baldwin 2003). Upland stream beds in South Otago, New Zealand, are characterised by sustained periods of desiccation and inundation. With time, 1-m-deep sediments have developed via water erosion from steep (commonly 10–20%) hill slopes. Recently, land use has changed from sheep farming to intensive dairy farming and, with it, nutrient losses via dung, urine and inorganic fertilisers have increased (McDowell et al. 2004). As a result, © CSIRO 2009

there is concern that stream productivity will be changed and accelerated eutrophication will occur. Knowledge of sediment P dynamics will aid in impact assessment. Hence, my primary hypothesis was that sediments in dairy-farmed catchments were enriched in bioavailable P compared with those in sheep-farmed catchments in South Otago, New Zealand, and that the concentration of bioavailable P was enhanced in dry compared with wet conditions. Materials and methods Site details and sampling Sites were located within the South Otago rolling downlands, within 50 km of Balclutha, New Zealand (5436960N, 2257880E). Catchments were selected with the following characteristics: 1st-order streams only; land use had remained constant for at least 5 years (negating the possible effect of P losses associated with capital fertiliser applications in the first year of operation); slope between 20% and 10%; elevation between 250 and 320 m above sea level; rainfall between 900 and 1000 mm annually; mean annual temperature between 8.8 and 9.2◦ C; stock had been kept out of the stream bed for the previous 6 months and would be fenced out during the study; and, soil type was a Waitahuna silt loam (NZ Classification, Mottled Fragic Pallic soil; USDA Taxonomy, Haplostoll; Otago Regional Council 2004). These conditions commonly created summer-dry sediments, with no flow between December and April and wet sediments at other times of the year, commonly yielding a low base flow (10 L s−1 ). 10.1071/MF08047

1323-1650/09/070619

620

Marine and Freshwater Research

Steep topography within catchments results in an extensive layer of sediment 1 m deep. A combination of wind and water erosion has commonly relocated grass seed from pasture to sediment and mixed grasses may establish during late spring. Plant growth on sediment typically occurs for 2 months, until sediment dries below permanent wilting point. Drying continues down to ∼30-cm depth by April. At this time, sediment moisture content is commonly 13 by taking 0.2 g of the freeze-dried extract, adding 600 µL of D2 O and 100 µL of 10 m NaOH. Samples were placed in an ultrasonic bath for 3 min, left to sit for 20 min then centrifuged (6-place mini-centrifuge) for 5 min. The supernatant was transferred to a 5-mm NMR tube and 31 P NMR spectra obtained at 202.298 MHz at 20◦ C. Accumulation of data for each sample was halted when a sufficient signal to noise ratio (>150) was obtained (4–12 h, 4800–16 200 scans). Scans were accumulated by using a pulse angle of 45◦ , a pulse delay of 4 s and an acquisition time of 1.99 s with 64 K data points. Chemical shifts were recorded relative to an external phosphoric acid standard (δ = 0 ppm) in a capillary tube. Spectra were deconvoluted by using a Lorentzian line shape of 5 Hz and measured with Mestre-C software (Gómez and López 2004; Mestrelab Research SL., Santiago de Compostela, Spain). Peak assignments were made quantitative by combining the percentage spectral area occupied by a peak and the total P concentration in the corresponding NaOH–EDTA extract. Peaks corresponding to classes of P compounds were taken from the literature (Newman and Tate 1980; Turner et al. 2003) and classified as orthophosphate, orthophosphate monoesters, orthophosphate diesters (DNA and phospholipids–techoic acid–P), pyrophosphates and phosphonates. For orthophosphate monoesters, peaks were assigned and deconvoluted for quantification of phytate by using peak assignments for the myo-inositol hexakisphosphate isomer (Turner et al. 2003). Other analyses and statistics All sediments were air-dried, crushed and sieved (20 ppm) did not change with

Fe(OOH)≈P CaCO3≈P ASOP NaOH-P ROP Total P

l.s.d.05 – Land use

800

l.s.d.05 – Moisture

1200

1000 600 800

400

600

400 200

Total P concentration (mg kg⫺1)

Land use Dairy Dairy Sheep Sheep l.s.d.land use l.s.d.moisture l.s.d.interaction

Moisture status

P concentration (mg kg⫺1)

Parameter

200

0

0 Dairy wet

Dairy dry

621

Sheep wet

Sheep dry

Land use and moisture status Fig. 1. Mean concentration (mg kg−1 ) and least significant difference (l.s.d.) at the P = 0.05 level of significance for phosphorus (P) fractions in wet and dry sediments from streams in dairy- and sheep-farmed catchments. Note that the mean total P concentration and corresponding l.s.d. are on a different y-axis. The interaction between land use and moisture was not significant and, hence, the corresponding l.s.d. values are not presented for brevity.

622


R. W. McDowell

Table 2. Mean concentration (mg kg−1 ), percentage in parentheses and least significant difference (l.s.d.) at the P = 0.05 level of significance for phosphorus forms in resolubilised NaOH–EDTA extracts of wet and dry sediments from streams in dairy- and sheep-farmed catchments The chemical shift (δ, ppm) of each peak assignment is given in parentheses Parameter

Moisture status

Land use Dairy Dairy Sheep Sheep l.s.d.land use l.s.d.moisture l.s.d.interaction A This

Ortho phosphate (∼6.2)

Monoesters (6 to 3)

Phytate (6 to 4)A

Diesters (2 to −1)

Pyro phosphate (−3 to −4)

Phos phonates (20)

183 (26) 282 (40) 148 (23) 265 (39) 34 67 151

475 (64) 376 (56) 452 (68) 375 (59) 34 37 94

186 (26) 174 (25) 183 (29) 172 (25) 46 44 52

39 (6) 18 (3) 29 (5) 12 (2) 12 16 31

8 (1) 31 (4) – 21 (3) 16 20 39

– – 2 (1) 2 (1) – 3 –

Wet Dry Wet Dry

monoester was determined from the assignment of myo-inositol hexakisphosphate (Turner et al. 2003).

Orthophosphate

Dairy – wet

Sheep – wet

Monoesters

Diesters

Pyrophosphate Phosphonates

Sheep – dry

Dairy – dry

20.0 ppm (t1) Fig. 2.

31 P

15.0

10.0

5.0

0.0

20.0 ppm (t1)

15.0

10.0

5.0

0.0

NMR spectra of NaOH–EDTA extracts of wet and dry sediments from representative dairy- and sheep-farmed catchments.

the moisture status of the sediment and they were found only in the sediments from sheep-farmed catchments. Discussion Sediment fractionation More P was found in the sediment of streams in dairy-farmed than in sheep-farmed catchments (Fig. 1). This was due to the greater Fe(OOH)≈P, CaCO3 ≈P and ASOP concentrations in the sediments of streams in dairy-farmed catchments. Much data exist to show that intensive dairy farms lose more P to streams

than do sheep farms because of a combination of increased P-fertiliser inputs, treading damage to the soil enhancing P loss by overland flow, management and animal behaviour (Gillingham and Thorrold 2000; McDowell et al. 2004). For instance, sheep tend to stay out of waterways, whereas dairy cattle will use them for shade, shelter and drinking (Byers et al. 2005). Furthermore, many farmers in South Otago put cattle into stream beds to utilise grass grown on stream banks during spring and early summer. This results in direct P inputs via dung deposition, and via mixing of dung with sediment by treading. Cattle dung is near neutral or slightly alkaline (Haynes and Williams

Land use and phosphorus in upland stream beds

1993) and, as a consequence, P can be expected to accumulate as CaCO3 ≈P, much like in soils (Hedley et al. 1982). Overall, total P concentration did not vary with the sediment moisture status, although P fractions did. Fractions that decreased in dry compared with wet sediments were CaCO3 ≈P and the two organic P fractions, ASOP and NaOH-P, whereas ROP and Fe(OOH)≈P increased. The decrease in CaCO3 ≈P can be explained by a decrease in pH (De Groot and Fabre 1993) or biological uptake (De Graaf Bierbrauwer-Würtz and Golterman 1989). Part of this decrease could also be associated with the mineralisation of organic P associated with CaCO3 ≈P. Other studies have found 5–20% of this fraction to be organic P (e.g. Kassila 2003). However, because the organic P component of P fractions was not measured in the present study, this fraction remains unknown. An increase in Fe(OOH)≈P has been observed for submerged sediments that have either been drained or allowed to dry (De Groot and Van Wijck 1993; Kassila 2003). De Groot and Van Wijck (1993) showed that when anoxic sediments were exposed to air, ferrous sulfides were rapidly oxidised to amorphous ferric (oxy)hydroxides. These have a much larger surface area and affinity for P than do crystalline Fe minerals. Fabre (1992) showed that more P was present in the NaOH-P fraction, being largely attributed to increases in Fe-P as the sediment dried. However, Baldwin (1996) showed that sediments above the oxycline (i.e. aerobic most of the time) decreased their sorption capacity and affinity for P when dried, owing to increasing crystallinity and mineral aging. Similarly, McDowell and Sharpley (2001) showed that with time, wetting and drying cycles shifted P into more unavailable forms, a phenomenon possibly expressed here by an increase in ROP. As a percentage of total P, concentrations of organic P forms were lower in dry than in wet sediment (Fig. 1), especially ASOP. In a laboratory experiment, De Groot and Van Wijck (1993) found that ASOP in a homogenised sediment suspension decreased from 260 mg kg−1 to 186 mg kg−1 when allowed to dry for 27 days. Furthermore, De Groot and Fabre (1993) found that ASOP was specifically mineralised by ∼40% during desiccation of a freshwater marsh. They concluded that this fraction was bioavailable; the present data support this, although they are not conclusive because the exact composition of this fraction is unknown. In contrast, NaOH-P is known to contain a large amount of humic material and phytate (De Groot and Golterman 1993). Although this fraction decreased only 5% in the top two sediment layers between samplings, the difference was significant indicating some of this P was possibly mineralised and/or bioavailable. 31 P NMR can determine whether this was due to phytate or humic material. 31 P

NMR To investigate the changes in organic P fractions, sediments were extracted with NaOH–EDTA and total P was determined. With the addition of EDTA, NaOH extraction of organic P is maximised by the chelation of transition metals (Bowman and Moir 1993). Consequently, because NaOH–EDTA will extract P from many pools, this fraction is not comparable to the NaOH-P fraction in the Golterman fractionation scheme. However, the present data clearly show that (1) the extraction of organic P, and indeed total P, is efficient and (2) more P is extracted from dry than from wet sediments. This suggests that, like fractionation data,


623

NaOH–EDTA is a sensitive measure of changes in the forms and concentrations of P in the sediment. Investigation of the P forms in NaOH–EDTA extracts by 31 P NMR indicated that orthophosphate was the major P form. More orthophosphate was present in dry than wet sediments and in sediments from dairy- than sheep-farmed catchments (Table 2). The difference with land use can be explained by the dominance of orthophosphate in dairy-cattle dung (McDowell and Stewart 2005) and by earlier dung deposition on the stream bed. McDowell and Stewart (2005) also demonstrated that orthophosphate occupied 84% of the total P in the spectra of NaOH–EDTA extracts of dry dairy-cattle dung and only 73% in the extracts of wet dung because of decomposition of organic P compounds into base constituents, including orthophosphate. The increase in orthophosphate extracted from sediments was about equal to the decrease in diester P, pyrophosphate and monoester P other than phytate (Table 2). Apart from orthophosphate, other inorganic P compounds usually found in sediments and soils by 31 P NMR are polyphosphate and pyrophosphate (the smallest version of a polyphosphate). No polyphosphates were found in either wet or dry sediments. Possible reasons for this include degradation during extraction (Turner et al. 2003) or no luxury uptake and storage by microbes (Khoshmanesh et al. 2002). Although degradation of pyrophosphates was noted in alkaline NaOH extracts by Hupfer and Gächter (1995), all extracts had the same extraction time and potential for degradation. The exact reasons for the differences in pyrophosphate are unclear. However, increased P fertilisation and dung inputs with entrained pyrophosphate (McDowell and Stewart 2005) may have artificially increased pyrophosphate in the sediments from dairy-farmed catchments compared with those from sheep-farmed catchments. Several authors have also shown that pyrophosphate is metabolised under anaerobic conditions in wet sediments (e.g. Sundareshwar et al. 2001), whereas Makarov et al. (2005) showed that dry aerobic conditions promote fungal growth which stores pyrophosphate. Of the organic P species, most were orthophosphate monoesters, followed by orthophosphate diesters and phosphonates. Diesters could be separated into DNA and phospholipids. However, peaks were too broad to quantify them separately and accurately. In general, the presence of diesters can be used as an indicator of biological activity. For instance, Watts et al. (2002) found that phospholipid concentration could be used as a surrogate measure of changes in microbes, and many authors have suggested that organic P turnover in soils occurs largely via diesters (e.g. Turner et al. 2003). In the samples of the present study, the concentration of diesters was lower in dry than in wet sediments. Carman et al. (2000) showed that utilisation of diesters was more efficient in oxic than anoxic conditions. However, it is impossible to say definitively that this was due to biotic activity because diesters are known to degrade rapidly (Turner et al. 2003). It is interesting to note that the concentration of phytate did not change much among sediment samples with changing moisture regime. Phytate comprises a large proportion of organic P in soils and sediments (White and Miller 1976; Stevenson 1986). With up to six orthophosphate moieties, phytate can bind more strongly to sediment constituents such as Al and Fe oxides than to a single orthophosphate or most other organic P compounds

624


(Leytem et al. 2002). Golterman et al. (1997) hypothesised that phytate would bind to Fe(OOH) produced in summer sediment by oxidation of FeS. This can impart physical protection to phytate from biological attack (Kassila 2003). Furthermore, inositol phosphates are resistant to acid and base hydrolysis (White and Miller 1976; Stevenson 1986), meaning they are very stable in NaOH–EDTA extracts. Phosphonates (>20 ppm) are also thought to be recalcitrant (Newman and Tate 1980) and did not change with moisture status. However, they were found only in sediments from sheep catchments. Because they constituted diesters > pyrophosphate > monoesters (including phytate) > phosphonates. Coupling fractionation and NMR data together indicated more bioavailable P is present in sediments from dairy-farmed catchments than in sediments from sheep-farmed catchments, especially when dry. This suggests that substantial P reserves may act as a source of P for in-stream productivity, even if no further P inputs occurred. This was especially the case for dairy-farmed catchments, presumably owing to increased inputs compared with sheep-farmed catchment. To improve or maintain water quality in these streams, additional management such as dredging may be required. Acknowledgements I thank Ian Stewart of the Department of Chemistry, University of Otago, New Zealand, for generating the spectra. Funding for this work was provided by the New Zealand Foundation for Research Science and Technology under contract C10X0320. The paper was enhanced by the suggestions of both reviewers and the guest editor.

References Baldwin, D. S. (1996). Effects of exposure to air and subsequent drying on the phosphate sorption characteristics of sediments from a eutrophic reservoir. Limnology and Oceanography 41, 1725–1732.

R. W. McDowell

Bedrock, C. N., Cheshire, M. V., Chudek, J. A., Goodman, B. A., and Shand, C. A. (1994). Use of 31 P-NMR to study the forms of phosphorus in peat soils. Science of the Total Environment 152, 1–8. doi:10.1016/ 0048-9697(94)90545-2 Bowman, R. A., and Moir, J. O. (1993). Basic EDTA as an extractant for soil organic phosphorus. Soil Science Society of America Journal 57, 1516–1518. Byers, H. L., Cabrera, M. L., Matthews, M. K., Franklin, D. H., Andrae, J. G., Radcliffe, D. E., McCann, M. A., Kuykendall, H. A., Hoveland, C. S., and Calvert, V. H. (2005). Phosphorus, sediment, and Escherichia coli loads in unfenced streams of the Georgia Piedmont, USA. Journal of Environmental Quality 34, 2293–2300. doi:10.2134/JEQ2004.0335 Carman, R., Edlund, G., and Damberg, C. (2000). Distribution of organic and inorganic phosphorus compounds in marine and lacustrine sediments: a 31 P NMR study. Chemical Geology 163, 101–114. doi:10.1016/S00092541(99)00098-4 De Graaf Bierbrauwer-Würtz, I. M., and Golterman, H. L. (1989). Fosfaatfracties in de bodem van een aantal Nederlanse meren. H2 O 22, 411–414. De Groot, C. J., and Fabre, A. C. (1993). The impact of desiccation of a freshwater marsh (Garcines Nord, Carmague, France) on the sediment– water–vegetation interactions. Part 3: the fractional composition and the phosphate adsorption characteristics of the sediment. Hydrobiologia 252, 105–116. De Groot, C. J., and Golterman, H. L. (1993). On the presence of organic phosphate in some Camargue sediments: evidence for the importance of phytate. Hydrobiologia 252, 117–126. De Groot, C. J., and Van Wijck, C. (1993). The impact of desiccation of a freshwater marsh (Garcines Nord, Camargue, France) on sediment–water–vegetation interactions. Part 1: The sediment chemistry. Hydrobiologia 252, 83–94. Fabre, A. C. (1992). Inorganic phosphate in exposed sediments of the River Garonne. Hydrobiologia 228, 37–42. doi:10.1007/BF00006474 Fierer, N., Schimel, J. P., and Holden, P. A. (2003). Influence of drying– rewetting frequency on soil bacterial community structure. Microbial Ecology 45, 63–71. doi:10.1007/S00248-002-1007-2 Gillingham, A. G., and Thorrold, B. S. (2000). A review of New Zealand research measuring phosphorus in runoff from pasture. Journal of Environmental Quality 29, 88–96. Golterman, H. L. (1996). Fractionation of sediment phosphate with chelating compounds: a simplification, and comparison with other methods. Hydrobiologia 335, 87–95. doi:10.1007/BF00013687 Golterman, H., Paing, J., Serrano, L., and Gomez, E. (1997). Presence of and phosphate release from polyphosphate or phytate phosphate in lake sediments. Hydrobiologia 364, 99–104. doi:10.1023/ A:1003212908511 Gómez, J. C. C., and López, F. J. S. (2004). Mestre-C, Nuclear Magnetic Resonance Companion. v. 3.5.1. Available at http://www.mestrec.com [verified November 2008]. Haynes, J. R., and Williams, P. H. (1993). Nutrient cycling and soil fertility in the grazed pasture ecosystem. Advances in Agronomy 49, 119–199. doi:10.1016/S0065-2113(08)60794-4 Hedley, M. J., Stewart, J. W. B., and Chauhan, B. S. (1982). Changes in inorganic and organic soil phosphorus fractions induced by cultivated practices and laboratory incubations. Soil Science Society of America Journal 46, 970–976. Howell, J. R., Donahoe, R. J., Roden, E. E., and Ferris, F. G. (1998). Effects of microbial iron oxide reduction on pH and alkalinity in anaerobic bicarbonate-buffered media: implications for metal mobility. Mineralogical Magazine 62A, 657–658. doi:10.1180/MINMAG.1998. 62A.2.13 Humphries, P., and Baldwin, D. S. (2003). Drought and aquatic ecosystems: an introduction. Freshwater Biology 48, 1141–1146. doi:10.1046/ J.1365-2427.2003.01092.X

Land use and phosphorus in upland stream beds


Hupfer, M., and Gächter, R. (1995). Polyphosphate in lake sediments: 31 P NMR spectroscopy as a tool for its identification. Limnology and Oceanography 40, 610–617. Kassila, J. (2003). Effects of lining and drying on the chemical composition of sediments and influence of organic carbon on carp growth in aquaculture ponds. Aquaculture and Research 34, 333–343. doi:10.1046/J.13652109.2003.00822.X Khoshmanesh, A., Hart, B. T., Duncan, A., and Beckett, R. (2002). Luxury uptake of phosphorus by sediment bacteria. Water Research 36, 774–778. doi:10.1016/S0043-1354(01)00272-X Kittredge, J. S., and Roberts, E. (1969).A carbon–phosphorus bond in nature. Science 164, 37–42. doi:10.1126/SCIENCE.164.3875.37 Leytem, A. B., Mikkelsen, R. L., and Gilliam, J. W. (2002). Sorption of organic phosphorus compounds in Atlantic coastal plain soils. Soil Science 167, 652–658. doi:10.1097/00010694-200210000-00003 Makarov, M. I., Haumaier, L., Zech, W., Marfenina, O. E., and Lysak, L. V. (2005). Can 31 P NMR spectroscopy be used to indicate the origins of soil organic phosphates? Soil Biology & Biochemistry 37, 15–25. doi:10.1016/J.SOILBIO.2004.07.022 McDowell, R. W., and Sharpley, A. N. (2001). A comparison of fluvial sediment phosphorus (P) chemistry in relation to location and potential to influence stream P concentrations. Aquatic Geochemistry 7, 255–265. doi:10.1023/A:1015274426411 McDowell, R. W., and Stewart, I. (2005). Phosphorus in fresh and dry dung of grazing dairy cattle, deer and sheep: Sequential fractionation and 31 P nuclear magnetic resonance. Journal of Environmental Quality 34, 598–607. McDowell, R. W., Biggs, B. J. F., Sharpley, A. N., and Nguyen, L. (2004). Connecting phosphorus loss from agricultural landscapes to surface water quality. Chemistry and Ecology 20, 1–40. doi:10.1080/ 02757540310001626092 Newman, R. H., and Tate, K. R. (1980). Soil phosphorus characterisation by 31 P nuclear magnetic resonance. Communications in Soil Science and Plant Analysis 11, 835–842. doi:10.1080/00103628009367083 Otago Regional Council (2004). GrowOtago climate and soils maps. Otago Regional Council Limited, Dunedin, New Zealand. Available from http://www.govt.orc.nz/ [Accessed January 2008].

625

Paytan, A., Cade-Menun, B. J., McLaughlin, K., and Faul, K. L. (2003). Selective phosphorus regeneration of sinking marine particles: evidence from 31 P-NMR. Marine Chemistry 82, 55–70. doi:10.1016/S03044203(03)00052-5 Qiu, S., and McComb, A. J. (1994). Effects of oxygen concentration on phosphorus release from reflooded air-dried wetland sediments. Australian Journal of Marine and Freshwater Research 45, 1319–1328. doi:10.1071/MF9941319 Qiu, S., and McComb, A. J. (1995). Planktonic and microbial contributions to phosphorus release from fresh and air-dried sediments. Marine and Freshwater Research 46, 1039–1045. doi:10.1071/MF9951039 Sah, R. N., Mikkelsen, D. S., and Hafez, A. A. (1989). Phosphorus behaviour in flooded-drained soils. 2. Iron transformations and phosphorus sorption. Soil Science Society of America Journal 53, 1723–1729. Stevenson, F. J. (1986). ‘Cycles of Soil.’ (J. Wiley and Sons: New York.) Sundareshwar, P. V., Morris, J. T., Pellechia, P. J., Cohn, H. J., Porter, D. E., and Jones, B. C. (2001). Occurrence and ecological implications of pyrophosphate in estuaries. Limnology and Oceanography 46, 1570–1577. Tate, K. R., and Newman, R. H. (1982). Phosphorus fractions of a climosequence of soils in New Zealand tussock grassland. Soil Biology & Biochemistry 14, 191–196. doi:10.1016/0038-0717(82)90022-0 Turner, B. L., Mahieu, N., and Condron, L. M. (2003). The phosphorus composition of temperate pasture soils determined by NaOH-EDTA extraction and solution 31 P NMR spectroscopy. Organic Geochemistry 34, 1199–1210. doi:10.1016/S0146-6380(03)00061-5 Watts, E. E., Dean, P. A. W., and Martin, R. R. (2002). 31 P nuclear magnetic resonance study of sediment microbial phospholipids. Canadian Journal of Analytical Cciences and Spectroscopy 47, 127–133. White, R. H., and Miller, S. J. (1976). Inositol isomers: occurrence in marine sediments. Science 193, 885–886. doi:10.1126/SCIENCE.193.4256.885

Manuscript received 23 February 2008, accepted 26 November 2008

http://www.publish.csiro.au/journals/mfr