| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Finnish Environment Institute, P.O. Box 140, FIN-00251 Helsinki, Finland
* Corresponding author (risto.uusitalo{at}mtt.fi)
Received for publication April 8, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AER, anion exchange resin • DCAA, dual culture algal assay • DL, detection limit • DRP, dissolved molybdate-reactive phosphorus • PP, particulate phosphorus • TDP, total dissolved phosphorus • TP, total phosphorus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Particulate P is typically the dominant physical P fraction in turbid runoff from cultivated clayey soils (Pietiläinen and Rekolainen, 1991; Turtola and Paajanen, 1995). Whereas the potentially algal-available portion of PP consists mainly of inorganic non-apatite P (Williams et al., 1980), soils and eroded sediment may also contain a large reserve of sparsely soluble P associations (Kaila, 1964; Hartikainen, 1979), for example, apatite. Then, the total phosphorus (TP) flux transported by agricultural runoff may be a poor proxy for the eutrophying P load (Ekholm, 1998).
The potential availability of PP in runoff has been estimated by various types of algal assays and chemical methods. Algal assays are generally considered the more accurate option (Hegemann et al., 1983; Boström et al., 1988), but they are expensive and lengthy. Thus, the relatively simple chemical methods, for example, extraction with 0.1 M NaOH (Cowen and Lee, 1976; DePinto et al., 1981; Dorich et al., 1985) and with P sinks, such as ion exchangers (Cowen and Lee, 1976; Huettl et al., 1979; Hanna, 1989) and iron oxide (FeO)–impregnated filter paper (Ekholm and Yli-Halla, 1992; Sharpley, 1993), are a tempting alternative. Phosphorus sinks keep the solution P concentration low and thereby enhance desorption of PP, which mimics the action of growing algae.
The amount of P retained by P sinks has been correlated with the results obtained by algal assays. Huettl et al. (1979) equilibrated sediment suspensions with hydroxy-Al saturated cation exchange resin and found that the P retained by the resin correlated (R2 = 0.98; n = 5) with P uptake by Selenastrum capricornutum. For runoff sediment from agricultural land, Sharpley (1993), in turn, showed that P extracted with FeO paper strips correlated (R2 = 0.92–0.96) with the number of cells of several freshwater algal species in batch experiments. As a pretreatment, these authors concentrated runoff sediment, and used slurries with fixed sediment to solution ratios.
When analyzing water samples containing variable amounts of suspended sediment, however, Ekholm and Yli-Halla (1992) found that P uptake by Selenastrum capricornutum in a dual culture algal assay (DCAA) correlated poorly with FeO-P. Iron oxide strips seemed to extract more P from samples rich in suspended sediment than did the DCAA, whereas in the samples with a low sediment concentration the opposite occurred. The authors concluded that, as well as desorbed PP, the suspended soil particles themselves had attached to the FeO strips. This resulted in overestimation of the reversibly adsorbed PP in the samples rich in sediment. Thus, the FeO strip method may include an error source that is difficult to control in analyses of runoff samples with variable amounts of suspended soil. Such an artifact should be less important in extraction with anion exchange resin (AER; Uusitalo and Yli-Halla, 1999).
Using a direct P analysis of river water and of municipal and industrial wastewater, Hanna (1989) found a close correlation (R2 = 0.83; n = 49) between the concentration of algal-available P and AER-extractable P, on average 92% of the algal-available P being extractable with AER. In his study, the samples were continuously repumped for 24 h through an AER column packed with Duolite A 101D strongly basic AER (Rohm and Haas, Philadelphia, PA) in a OH- form. Later, Uusitalo et al. (2000) extracted P directly from agricultural runoff samples with a modification of the procedure of Sibbesen (1977)(1978), in which Dowex 1 x 8 strongly basic AER (Dow, Midland, MI) is enclosed in mesh bags and converted into a HCO-3 form. This procedure is simpler than that of Hanna (1989) and allows rapid determination of AER-extractable P; it has not, however, been compared with an algal assay.
The estimated prime cost of the AER bag procedure is less than 10 Euros per sample (1 Euro = about US $0.9–1.0), as compared with about 570 Euros per sample (in duplicated analysis) for the DCAA of Ekholm (1994). Although batch assays are somewhat cheaper than the DCAA, large surveys are probably not feasible with any type of algal assay. In the search for a complementary method for algal assays, we (i) compared the results of the AER test and the DCAA, primarily to be able to relate the amounts of AER-P and DCAA-P in turbid agricultural runoff, and (ii) investigated the performance of these methods.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
To test whether the AER extracts the same P pool as is utilized by the test algae, two of the turbid field runoff samples (Aurajoki composite and Jokioinen composite; Table 1) were (in addition to separate AER and DCAA analyses) first extracted by the AER method and then the same subsample was analyzed by the DCAA. Additional samples, used to test the performance of the AER method but not tested by the DCAA, are not listed in Table 1, but were collected from the three Cryaquepts mentioned above.
Analyses for Phosphorus and Suspended Solids
All undiluted samples were analyzed for their concentrations of dissolved molybdate-reactive phosphorus (DRP) and total phosphorus (TP) without replicates; these results are presented in Table 1. Then, after the samples were diluted for the DCAA, duplicated DRP analysis and quadruplicated TP analysis were conducted. For the determination of DRP, an aliquot was passed through a 0.4-µm Nuclepore polycarbonate filter (Whatman International, Maidstone, UK) and for the determination of TP, unfiltered aliquots were digested with K2S2O8 and H2SO4 in an autoclave at 120°C and 100 kPa for 30 min. Phosphorus concentrations were measured by the molybdenum blue method (Murphy and Riley, 1962) at 880 nm with a Hitachi (Tokyo, Japan) U-2000 spectrophotometer equipped with a 5-cm path cell. Suspended solids (SS) were determined by weighing the amount of dried (105°C) matter retained by the 0.4-µm Nuclepore filters (replicated in the same manner as TP).
Anion Exchange Resin Extraction
Anion exchange resin bags were prepared by weighing 1 g of Dowex 1 x 8 strongly basic AER into nylon mesh bags (netting mesh 0.25–0.30 mm; Sefar Inc., Heiden, Switzerland), as described by Sibbesen (1977). The total anion exchange capacity of one AER bag was about 2 mmolc. Before use, the AER was converted into a HCO-3 form (Sibbesen, 1978).
In the AER analysis, an undiluted 40-mL sample was shaken with one AER bag for 20 h on an end-over-end shaker at 37 rpm. The AER bag was removed from the sample, washed with deionized water, and shaken for 4 h in 40 mL of 0.5 M NaCl to displace P from the AER into the solution. The NaCl solution was then acidified with 1 mL of 6 M HCl and allowed to stand overnight. A Lachat (Milwaukee, WI) QC autoanalyzer was used to measure the P concentration of the NaCl solution. All AER extractions were performed in triplicate, unless otherwise specified. Each test series included one blank sample (deionized water) and two P standards (of K2HPO4 in deionized water).
Dual Culture Algal Assay
In the DCAA, a P-starved culture of Selenastrum capricornutum Printz and a sample (both 275 mL) were incubated on a shaking table at 20 ± 1°C and at 4200 ± 200 lux for 3 wk (occasionally 2 wk). The incubation took place in a two-chambered vessel, in which the sample and the algal suspension were separated by a 0.4-µm Nuclepore membrane. The sample chambers were covered with aluminum foil to reduce light penetration and growth of indigenous algae, and the algal chambers were aerated to prevent CO2 deficiency. The system was buffered at pH 8 with tris(hydroxymethyl)-aminomethane (Trizma Base; Sigma-Aldrich, St. Louis, MO).
After each week of incubation, the algal cultures were replaced with fresh P-starved cultures to ensure P limitation. Both fresh and harvested algal suspensions were filtered (0.4 µm) and analyzed for DRP within a day of the harvest. Total dissolved phosphorus (TDP) in the 0.4-µm filtrate and TP of the algal suspension were analyzed within two days of the harvest with the autoclave-mediated digestion described earlier. The difference between TP and TDP was assumed to represent P in algae, and the algal-available P in a sample was taken as the sum of the amount of P taken up by the consecutive cultures. For a more detailed discussion of the method, see DePinto et al. (1981) and Ekholm (1994)(1998).
Whenever the samples contained more than about 200 µg L-1 of TP, they were diluted with P-free nutrient medium before testing. The greatest dilutions were 10- and 250-fold for the field runoff and wastewater samples, respectively. The assays were performed in duplicate or triplicate. Blank samples with no P (i.e., P-free nutrient medium as a sample) were included, and treated as the actual samples. Each test series also contained one to three control vessels consisting of nutrient medium with 20 to 108 (mostly 72) µg P L-1; the treatment of these control samples only differed from that of the actual samples in that the nutrient medium in the sample chamber was spiked with P in each harvest. At the end of the test, an aliquot was taken from the sample chamber and analyzed for DRP, TDP, and TP. This permitted calculation of the P balance of the system (the ratio of P added to the vessel to the P recovered from it), serving as an indication of the overall accuracy of the test procedure.
The AER and DCAA analyses were conducted separately at two laboratories specializing in these methods: the AER analyses at the laboratory of MTT Agrifood Research Finland and the DCAA analyses at the laboratory of Pirkanmaa Regional Environment Centre.
Performance of the Methods
The operating ranges of the AER and DCAA methods were examined by determining the minimum detectable value, that is, the detection limit (DL), and the linearity of the response to changes in the P concentration within the range usually found in runoff samples. Uncertainty in the analytical results was assessed by determining the systematic and random errors of the methods, analytical recovery, and repeatability. For the AER extraction, the effects of sample dilution and anion competition on the AER-P results were also tested. In turn, for the DCAA, one potential source of systematic error, the uptake of P by bacteria in the sample chambers, was examined by direct bacterial counts.
The DL was calculated as follows (CITAT and EURACHEM, 2002):
 | [1] |
For the AER extraction, the DL was calculated from 112 blanks included in the roughly 50 test series conducted in 1997–1999. The DLs for the AER bags used in less than 15 extractions and for those used in 20 to 30 extractions were calculated separately. For the DCAA method, the DL was calculated from 16 blanks analyzed in 12 separate test series.
The linearity of the response to the changes in the solution P concentration was tested by analyzing standard P solutions (K2HPO4) of 0 to 2000 µg P L-1 for the AER test, and 20 to 110 µg P L-1 for the DCAA; the dilution of P-rich samples explains the narrower range for the algal assay. For the AER procedure, standard P solutions were analyzed in four separate test series performed on different days; for the DCAA, they were analyzed in two separate test series.
The presence of additive and multiplicative systematic errors was studied by testing the intercept and slope values of the regression lines as suggested by Doerffel (1994). Here, the hypotheses tested were that the intercept deviates from the value 0 and that the slope deviates from the value 1. The coefficient of variation (CV) of replicated analyses was used as a measure of the random error of the methods. For the AER method, the results of the 151 samples with AER-P 0 to 200 µg L-1 (analyzed in 1998 and 1999, all with triplicates) and for the DCAA, the 145 samples with DCAA-P 0 to 170 µg L-1 (analyzed in 1992–2000, with two or three replicates) were used.
The analytical recovery was assessed with the aid of P solutions of known concentrations. For the AER test, this was done by studying (i) P recovery from standard solutions containing 100 and 400 µg P L-1 (analyzed in 19 and 25 separate test series and days, respectively), and (ii) recovery of P from eight runoff samples (103–421 µg AER-P L-1) spiked with 15.3 µg PO4–P (3 mL x 5.09 mg P L-1) 2 h before AER extraction. Phosphorus recovery in the DCAA was studied with the aid of the data of control vessels (33 in total).
The effect of sample dilution on the results of the AER extraction was checked with five runoff samples: Aurajoki 7 (13 Jan. 2000), Aurajoki 16 (26 Nov. 1999 and 13 Jan. 2000), and Aurajoki 18 (26 Nov. 1999 and 13 Jan. 2000). These were tested as undiluted and as diluted with deionized water using the same dilution factor (ranging from about 4 to 10) as in the DCAA.
In the AER extraction, other anions competing for AER sorption sites may affect P recovery (Skogley and Dobermann, 1996). The interference caused by Cl-, NO-3, and SO2-4 was tested by recording the P recovery from two-component standard solutions containing 300 µg P L-1 (12 µg P in a 40-mL sample) and different concentrations of a competing anion. The concentration of Cl- in the fortified solutions ranged from 0 to 180 mg L-1 (0–203 µmolc in a 40-mL sample), the concentration of NO-3 from 0 to 320 mg L-1 (0–206 µmolc in 40 mL), and that of SO2-4 from 0 to 700 mg L-1 (0–583 µmolc in 40 mL). The solutions were prepared with analytical grade NaCl, KNO3, and K2SO4. There were six replicates in the tests.
To interpret the practical significance of competing anions in the AER test, the concentrations of the above competing anions in six agricultural basins (Aurajoki, Haapajyrä, Hovi, Löytäneenoja, Savijoki, and Uskelanjoki) intensively monitored in 1990–2000, were retrieved from the water quality database of the Finnish Environment Institute. The area of the basins ranged from 0.12 to 890 km2 and the cover of arable land from 27 to 100%. Haapajyrä differs from the other basins in that it is characterized by acid sulfate soils.
Bacterial growth in the DCAA vessels was tested by studying five runoff, two wastewater, and three river water samples, one blank sample, and three P controls with 72 µg P L-1. The bacteria in the sample chambers were counted before and after the DCAA following staining with DAPI (4'6-diamidino-2-phenylindole; Porter and Feig, 1980).
Statistical Analyses
The SAS System for Windows 2000 (Release 8.01; SAS Institute, 2000) was used to calculate the regression equations and statistical tests presented in this paper. Statistical comparisons between the results obtained by the AER test and the DCAA were done (after square root transformation of the original data to normalize the residuals) using analysis of variance (ANOVA) for generalized randomized complete block design, with sample being a block factor in the model. The 14 turbid samples (i.e., field runoff) and the 11 samples containing little suspended matter (wastewater and water draining from agricultural and forested basins) were tested separately. The same procedure and transformations were used when testing whether dilution of five turbid runoff samples has a profound effect on the results of the AER extraction and its relation to DCAA-P. Wilcoxon signed rank test (e.g., Johnson, 1994) was used to test whether the amounts of P that were added and those recovered (analytical recovery) differed from each other. The repeatability of the methods, that is, possible changes in the results between test series, was estimated by one-way ANOVA of the results obtained in replicated analyses of P standards used as control samples.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
For the 14 field runoff samples from the Aurajoki, Jokioinen, and Sjökulla fields, in which TP consisted mainly of PP (Table 1), the results obtained by the two methods clearly differed from each other (p < 0.001). The DCAA gave higher P yields than did the AER test (Fig. 1) , but there was a strong correlation between the results. The P yield of the DCAA was also higher than AER-P for the samples from the agricultural basins, which likewise had PP as the major P form. For the 17 samples representing runoff from agricultural fields or basins having AER-P greater than the DL, AER-P corresponded to 72% (range 48–92%) of the 3-wk DCAA-P (R2 = 0.94; not shown) and to 84% (52–104%) of the 1-wk DCAA-P (R2 = 0.98; not shown). For the samples in which P was mainly in the DRP form, the methods gave essentially equal P yields (Table 1; p = 0.694). The runoff from the forested basins (and the sample from the River Narva) had low P concentrations, mostly below the DL of the AER test and in many cases also below that of the DCAA (Table 1).
|
Whether AER was able to promote a steep P gradient between solid and solution phase during an overnight extraction was verified by using a set of five runoff samples with 360 to 590 µg L-1 AER-extractable P. Their initial DRP concentration ranged from 270 to 370 µg P L-1, and it decreased in all cases to 2 to 3 µg L-1 (approaching the DL of the DRP determination) during a 20-h contact with AER. In the DCAA, the DRP concentration was usually less than 1 µg L-1 (the DL of the bioassay laboratory) when determined.
Performance of the Methods
The DL was higher for the AER test than for the DCAA (Table 2). In addition, the DL of the AER test increased with the number of times the bags were used in the extractions. Both methods produced a linear response to increasing P concentrations of standard solutions, with randomly scattered residuals (data not shown). For the AER test, the null hypothesis (H0) that slope = 1 was rejected (p < 0.001; Table 3), showing the presence of a systematic error component, but for the DCAA, the systematic error, if present, was not statistically significant. Comparison of the slope values and the standard errors (SE) of their estimates, however, shows that the difference in the test for significance of the slopes was greatly affected by the larger scatter in the results of the DCAA.
|
|
For both methods, the random error (in Fig. 2 expressed as CV) decreased with increasing P concentration. Generally the variation between replicates tended to be smaller for the AER test than for the DCAA. There was relatively little variation between the replicates in the AER-P yields that were greater than the DL of the method, CV seldom being more than 10%. For the DCAA, the optimal concentration range of measurement was 92 to 175 µg P L-1; here CV was less than 10%. At 20 to 100 µg DCAA-P L-1, CV was typically less than 20%, but it was clearly greater than that when DCAA-P was less than 20 µg L-1.
|
|
Effect of Competing Anions on Phosphorus Recovery with the Anion Exchange Resin Test
The sample matrix, too, played a role in P recovery with the AER test. Compared with a single-component P standard solution, at least 90% of added P was recovered in the solutions containing less than 10 µmolc NO-3, SO2-4, or Cl- in a 40-mL subsample (Fig. 3)
. Corresponding solution concentrations for the competing anions were 15.5 mg NO-3 L-1, 12 mg SO2-4 L-1, and 8.9 mg Cl- L-1. Tenfold higher amounts of these anions brought the relative P recovery down to approximately 80 to 85% of that of the single-component P standard. Here, the effect of the different anions seemed to play a minor role, and the mass action, that is, the total equivalent concentration of the competing anions, dictated the relative P recovery.
|
|
In the wastewater (mean 380 cells µL-1) and river water (mean 930 cells µL-1) samples tested, the changes in bacterial numbers were relatively small and, when observed, the numbers mainly increased. Fewer bacteria in the wastewater than in the river water were due to the dilution of these P-rich samples before testing. In conclusion, it seems improbable that bacterial P uptake in the sample chamber could have caused a major systematic error in the DCAA. However, staining occasionally revealed red nonbacterial organisms, probably indigenous algae of the samples. They were very few, but if present abundantly, they may have an effect on P balance. No test algae, nor signs of growth of fungi, were observed in the sample chambers.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Even though desorption of PP in the presence of a P sink, such as AER, is considered to occur in a similar manner to the PP release caused by uptake by algae, there are some distinct differences between the action of AER and algae that may explain the clearly higher P yield of the DCAA from runoff samples. Probably the most important difference is that in AER extraction, equilibria form between reversibly sorbed soil P and solution P, on the one hand, and solution P and AER-P, on the other. In contrast, algal P uptake in the close to optimal conditions of the DCAA test can be considered a non-equilibrium process. Although the difference in the DRP concentration of the remaining sample solution, as measured after AER and DCAA tests, seems negligible (2–3 and less than 1 µg L-1, respectively), it has been shown that when P concentration in water is small, even a small decrease in P concentration in the water phase may result in a relatively large quantity of P being released from soil (Yli-Halla et al., 2002). During the productive period, the DRP concentration in surface waters may be virtually zero due to algal P uptake, a condition that cannot be fully met by equilibrating a sample with AER. Algae are also capable of excreting substances that enhance the solubilization of, for instance, organic P reserves in situations of inorganic P limitation (e.g., Newman and Reddy, 1993); whether this happened during the DCAA was not tested. Other studies, however, indicate that algal assays may underestimate the P reserves ultimately available (Peters, 1981; Ekholm and Krogerus, 2002).
Despite being the most obvious difference between the AER test and the DCAA, the duration of the AER test as such probably has only a minor influence on the P yields in runoff analysis. We found that a 44- or a 72-h AER extraction (results not shown) hardly increased the total amount of P extracted from field runoff and wastewater samples (on average less than 4% more AER-P was extracted in these longer extractions; n = 17). Earlier, Cowen and Lee (1976) showed that extraction times extending for up to 26 d had only a small effect on AER-P as compared with 24-h extractions. This would fit the hypothesis that the three-phase equilibrium discussed above actually dictates the AER-P yield in runoff samples, and the small amount of P remaining in solution seems to effectively decrease further desorption of solid-phase P. In soil analysis, the situation is obviously different (e.g., Sibbesen, 1978; Bache and Ireland, 1980), but then the reserves of desorbable P in relation to the total exchange capacity of the AER are also larger, and reaching the equilibrium between solid- and solution-phase P seems to take longer.
Another difference between the methods was the sample pretreatment, with the AER extractions being performed on undiluted samples and the DCAA usually on diluted ones. The higher AER-P yields of the diluted samples were probably due to changes in the solution composition, perhaps accompanied by changes in the relative sizes of the competing P sinks. In addition to the effect of competing anions, also cation composition of the solution has been shown to affect the P yields in the AER tests (Curtin et al., 1987). Furthermore, a decrease in the ionic strength of the water–soil suspension facilitates P desorption from soil (Yli-Halla and Hartikainen, 1996). There is probably a big difference between the capabilities of soil and runoff sediment to release P, in that the sediment has already been somewhat depleted in sorbed P during the erosion process (Yli-Halla et al., 1995) and we might consider P-poor suspended sediment as a competing P sink in the AER extraction.
The median concentrations of the competing anions in the data set obtained from the national water quality database were relatively low. In addition, the amount and type of the competing anions seemed to have little effect on the AER-P yields over a wide concentration range (note the logarithmic scale in the x axis of Fig. 3), and we suggest that anion competition probably has a fairly constant effect in most runoff samples. The monitoring data, however, imply that periodical interferences in the AER test may be experienced in some basins (with acid sulfate, saline, or sodic soils) as a result of anion competition. This is especially likely, since HCO-3 was not taken into account here (a separate estimate of the HCO-3 concentration was not available, but it may be the most abundant anion present in natural waters). Similarly, organic anions may compete for AER sorption sites.
Considering the overall performance of the methods, the estimates obtained by the AER test and the DCAA showed clearly more variation than did those obtained by the methods used in routine water quality monitoring (DRP and TP): the DLs were relatively high (especially for AER), the recovery of added P varied, given both known (anion competition in the AER test) and unknown (disappearance of ortho-P from the DCAA) interferences, and the random error was relatively high within the measuring range applied (especially for the DCAA). However, the measures of water quality used, which are based on physical division of P into dissolved and particulate fractions, may be unsatisfactory in terms of the algal availability of runoff P (Ekholm and Krogerus, 2002); moreover, they may be inaccurate in terms of inclusion of colloidal PP in the DRP fraction (Hens and Merckx, 2002).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  R. W. McDowell and R. J. Wilcock Sources of Sediment and Phosphorus in Stream Flow of a Highly Productive Dairy Farmed Catchment J. Environ. Qual., March 1, 2007; 36(2): 540 - 548. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. W. McDowell and R. J. Wilcock Particulate Phosphorus Transport within Stream Flow of an Agricultural Catchment J. Environ. Qual., November 1, 2004; 33(6): 2111 - 2121. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Uusitalo, E. Turtola, M. Puustinen, M. Paasonen-Kivekas, and J. Uusi-Kamppa Contribution of Particulate Phosphorus to Runoff Phosphorus Bioavailability J. Environ. Qual., November 1, 2003; 32(6): 2007 - 2016. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
