| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a MTT Agrifood Research Finland, FIN-31600, Jokioinen, Finland
b Finnish Environment Inst., P.O. Box 140, FIN-00251, Helsinki, Finland
c Helsinki Univ. of Technology, Lab. of Water Resources Engineering, P.O. Box 5200, FIN-02015, Espoo, Finland
* Corresponding author (risto.uusitalo{at}mtt.fi).
Received for publication February 27, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AER-PP, anion exchange resin-extractable particulate phosphorus • BD-PP, bicarbonate-dithionite-extractable particulate phosphorus • DRP, dissolved molybdate-reactive phosphorus • PP, particulate phosphorus • TP, total phosphorus • TSS, total suspended solids
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
As a proxy for dissolved orthophosphate, DRP is considered readily available for uptake by biota, whereas only a fraction of PP may be transformed into a bioavailable form (Ekholm, 1994; Reynolds and Davies, 2001). The PP forms that may become bioavailable in receiving waters are assumed to include soluble P associations, desorbable P (attached to mineral and organic particulate matter), and redox-labile P (Ekholm, 1998; James et al., 2002). In contrast, biotests have shown that primary apatite-P is a poor P source for algae (Williams et al., 1980). Bioavailability of P associated with organic sediment matter is by some authors considered low (Williams et al., 1980), whereas others have found that organic P compounds that are stable in terrestrial soils, such as inositol phosphates, may relatively rapidly mineralize in marine environments (e.g., Suzumura and Kamatani, 1995).
In runoff from cultivated fields, PP often makes up a significant P fraction, both in surface runoff (Heathwaite and Johnes, 1996; Fleming and Cox, 2001) and in drainflow (Turtola and Paajanen, 1995; Laubel et al., 1999; Simard et al., 2000). The bioavailability of PP is likely to vary greatly from one area to another (see Ekholm, 1998) because geochemical properties of runoff sediment vary according to the source (Williams et al., 1980; Grobler and Silberbauer, 1985). In an aerobic water column, desorption of PP is an important reaction that increases the P pool utilized by primary producers. Hence, chemical extractions related to the size of the desorbable PP reserves have been suggested as alternatives to costly biotests when assessing PP bioavailability in water environments (e.g., Dorich et al., 1985; Hanna, 1989; Sharpley, 1993). Extraction with anion exchange resin (AER) is one option and, even though AER extractions typically give lower P yields than algal biotests, they have been shown to correlate well with P uptake in algal assays (Hanna, 1989; Uusitalo and Ekholm, 2003).
The redox effects on P release may also be significant. Sallade and Sims (1997) and Penn et al. (2000), for example, reported that anoxic sediments released much greater amounts of P into the ambient water than when in an oxic state. In freshwater sediments, redox reactions are often reversible, and P may be resorbed onto the sediment after reoxidation; nevertheless, Gunnars and Blomqvist (1997) showed that even though reoxidized freshwater sediments may resorb the P they released when in a reduced state, marine sediments do so less effectively, and significant amounts of P may remain in DRP form in the water column. To what extent PP in field runoff contributes to redox-labile P in the sediments of receiving waters is an open question. However, an indication of the potential for P solubilization from runoff sediment at low redox-potentials can be obtained when solubilization of redox-sensitive (mainly Fe-associated) PP is brought about by additions of bicarbonate and dithionite to runoff samples (Uusitalo and Turtola, 2003).
Different water bodies have different abilities to cope with nutrient inputs. Effective eutrophication control therefore calls for management protocols designed for specific catchments (Reynolds and Davies, 2001). Where PP loss dominates over DRP loss, a condition common in clayey soils, knowledge of the tendency of PP to become solubilized in different circumstances would facilitate selection of site-specific options for cost-effective catchment management. Here, we assessed the tendency of PP to become transformed into a bioavailable form by conducting chemical extractions of PP transported by runoff from arable clayey fields. Assessment of the chemical reactivity of sediment-associated P was based on extractions of runoff samples without sediment preconcentration, and two different estimates were produced: (i) desorption tendency of PP in an aerobic water column, and (ii) the potential for PP solubilization in an anaerobic sediment. The advantages of these extractions lie, above all, in their rapidity and relatively modest cost. Our main objective was to estimate the significance of PP fractions as a form of potentially bioavailable P transported from cultivated soils. We also discuss the importance of flow pathways and soil P status for the transport of different P forms.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
The experimental setup for the Aurajoki site has been described by Puustinen (1994), and that of the Jokioinen field by Turtola and Paajanen (1995). The Lintupaju field has been described by Uusi-Kämppä and Yläranta (1992), and the Sjökulla field by Paasonen-Kivekäs et al. (1999). The field plots selected for this study were those for which annual primary cultivation with a moldboard plow or a cultivator was done in the autumn. Barley (Hordeum vulgare L.), wheat (Triticum aestivum L.), or oat (Avena sativa L.) were grown in the fields during the study.
Runoff Measurement and Sampling
The study was launched in September 1997, and data from Aurajoki, Jokioinen, and Lintupaju were analyzed from that date until August 2001; each annual P loss was estimated starting on 1 September and ending on 31 August. At these three sites, runoff measurement and water sample collection, including both surface and subsurface runoff at Jokioinen and surface runoff only at Aurajoki and Lintupaju, were conducted with a tipping bucket arrangement. Annual losses were summed by multiplying the measured flow volumes and the P concentrations of the corresponding flow fractions. The number of surface runoff samples that formed the basis for the annual loss estimates varied between the study years from 28 to 64 at Aurajoki, from 96 to 171 at Jokioinen, and from 40 to 49 at Lintupaju. The annual Jokioinen drainflow P loss estimates were based on 384 to 592 individual runoff samples.
At Aurajoki, water sampling for the laboratory analyses, normally conducted twice a week, was interrupted from time to time in winter because of economical circumstances, and for this reason, water quality data are incomplete for the study period September 2000 to August 2001. In spring 1999, as a result of heavy snow accumulation on the plowed plot and flow from outside the plot (as a result of freezing of the catch ditches surrounding the field), runoff measured by dataloggers was for a long period unrealistically high, even exceeding annual precipitation. The P loss estimates for the spring months of 1999 were therefore based on analyses of two plots only. Also, at some other occasions, flow from outside the plots (i.e., when catch ditches flooded and water ran onto a field plot) was, in addition to observations in field, detected as much sharper hydrograph peaks as compared with simultaneous flow measurements at the other field plots (12 plots in total). The amount of actual runoff was then estimated by assuming that the flow from a suspect plot would have followed the long-term (10-yr) average difference as compared with the other plots at the field. Also in these occasions, water analyses were done on the composite samples (because flow had been through the plot) and used with the corrected runoff volumes in the P loss calculations. For the whole 4-yr study period, the flow volumes corrected by this manner differed 8% from those measured by dataloggers at the studied three Aurajoki plots.
The Sjökulla data comprised automated runoff volume measurements made with H.F. Jensen (Type PSL) pressure sensors (Majestic Electronics Ltd., Oxford, UK) at 15-min intervals, and manual grab surface runoff and manual or automated (EPIC 1011 portable water sampler; Buhler Montec Ltd., Manchester, UK) drainflow samples taken from v-notch weirs. Weirs identified as S2 (for surface runoff sampling) and D4 (for subsurface drainage sampling) were located within the same field segment, whereas Weir D3 (for subsurface drainage sampling) was located at the other segment (numbering of the sampling weirs is the same as in Paasonen-Kivekäs et al., 1999).
During the study, the goal was to visit the Sjökulla field whenever it rained, and if runoff occurred, it was sampled. This strategy resulted in 26 surface runoff (S2) and 45 drainflow (D3) grab samples. Automated sampling at the drainflow Weir D4 produced several subsurface runoff samples per day when runoff occurred, but these were averaged on a daily basis, and 95 daily averages were used in the P loss calculations. The calculation of P loss estimates at Sjökulla was only done for the days when runoff was sampled. Then, the concentrations measured for grab samples (S2 and D3), or daily averaged concentrations (D4), were multiplied by the daily runoff volumes; the drainflow volume per hectare drained field that was measured at Weir D3 was considered representative for the drainflow per hectare at Weir D4 as well. It was thus assumed that the measured P concentrations of the samples approximated the daily averages, but further extrapolation (outside the days of sampling) was not made. As the flow volume observations continued without interruption during 1998 at surface runoff Weir S2 and drainflow Weir D3, this period (1 Jan. 1998 through 31 Dec. 1998) was used to compare the losses of different P forms at Sjökulla.
Water Analyses
Dissolved molybdate-reactive P was determined after filtration of the samples through 0.2-µm or, for the Aurajoki runoff monitoring, 0.4-µm Nuclepore filters (Whatman, Maidstone, UK). Total P was determined after 30 min of autoclave-mediated digestion (120°C, 100 kPa, with K2S2O8 and H2SO4) of an unfiltered subsample, and the concentration of PP was taken as the difference between TP and DRP. Modification of the molybdenum blue method (Murphy and Riley, 1962) was employed in photometric P analysis. The concentration of total suspended solids (TSS) was estimated by weighing the evaporation residue of 40 to 80 mL of runoff, except at Aurajoki, where the mass of the dried (105°C) matter retained on the 0.4-µm Nuclepore filter was used as a proxy for TSS. The reason for the different DRP and TSS analyses for the Aurajoki samples, as compared with the other sites, was that the chemical analyses of the Aurajoki samples were done at the Southwest Finland's Regional Environment Centre (following the national standards for water analysis), whereas the other samples were analyzed at Agrifood Research Finland.
Assessment of desorbable PP was based on extraction of runoff with the AER procedure discussed by Uusitalo and Ekholm (2003). An undiluted 40-mL sample was shaken end-over-end at 37 rpm for 20 h with an AER bag containing 1 g of Dowex (Dow Chemicals, Midland, MI) 1 x 8 strongly basic AER. After the AER bag was removed from the sample and washed with deionized water, it was 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 before P determination by the molybdenum blue method. Desorbable particulate phosphorus (AER-PP) was taken as the difference between AER-extractable P and DRP. All samples analyzed for AER-P were also analyzed for DRP (with 0.2-µm filters), TP, and TSS (evaporation residue).
The number of samples per field for which the AER extraction was made ranged from 127 at Sjökulla to 213 at Jokioinen (Table 2). Sampling for AER analysis was done during 1998 to 2001 at Aurajoki, 1997 to 2001 at Jokioinen, 1997 and 1998 at Lintupaju, and 1997 to 1999 at Sjökulla. The AER analysis was always done with triplicates, mostly within a week of sampling [the amount of AER-extractable P was found to remain very constant for several weeks (Uusitalo, unpublished data, 1998)].
|
|
The extraction proceeded as follows: 40 mL of runoff was measured into a 50-mL-capacity centrifuge tube, to which were added 1 mL of 0.298 M NaHCO3 and 1 mL of 0.574 M Na2S2O4 solutions. The bicarbonate solution was prepared for daily use, whereas dithionite was dissolved for each extraction series separately just before it was added to the sample. After 15 min of shaking on an orbital shaker at 120 rpm, the sample was filtered through a 0.2-µm Nuclepore. For colorimetric P determination by the molybdenum blue method, an aliquot of the filtrate was digested in an autoclave with peroxodisulfate and sulfuric acid, as in TP analysis. From the P concentration of these digests, DRP was subtracted to give the concentration of BD-PP. All samples analyzed for BD-extractable P were also analyzed for DRP (passed through 0.2-µm filters), TP and TSS (evaporation residue). The BD-analysis was mostly performed without replicates to be able to construct BD-PP vs. PP relationships with as many independent observations as possible and with the laboratory resources available.
A separate analysis of total dissolved phosphorus (TDP; a subsample of 0.2-µm filtrate digested as in the TP analysis), that would also account for nonreactive P passing though the filter, and to be after the digestion step taken as BD-extractable, was not made for all samples. This was because the difference between DRP and TDP for a set of 49 samples was small in relation to BD-extractable P: the average difference between DRP and TDP in these samples was 0.018 mg L-1, as compared with 0.370 mg L-1 BD-P. The BD-PP estimates would then only be about 6% greater when DRP was used in the calculation instead of TDP.
Aurajoki runoff, which was sampled for BD-PP analysis during spring 2002, was analyzed in April 2002. The BD-PP analyses for the Jokioinen samples, which were taken during 1998, 2000, and 2001, and for the Sjökulla samples, which were taken during 1998 and 1999, were performed in May 2000 and June 2001. The Lintupaju field samples, collected in 1998, 2001, and 2002, were analyzed for BD-PP in June 2001 and April 2002.
The proxies for PP bioavailability, AER-PP and BD-PP, are taken as alternative estimates for different water body conditions rather than as additive values. In principal, the AER and BD extractions could also be conducted sequentially, but it was suspected that the sorption sites (e.g., Al oxides), which become P-depleted during AER extraction but do not respond to redox changes, could have captured much of the P solubilized during the subsequent BD extraction and thus suppressed the BD-PP yields.
Assessment of Desorbable and Redox-Sensitive Particulate Phosphorus Losses
Because we were not able to make AER- and BD-extractions for all of the nearly 3000 individual runoff samples taken from the four sites of this study, annual losses of desorbable particulate phosphorus (AER-PP) and redox-sensitive particulate phosphorus (BD-PP) were estimated with the aid of the relationships (Fig. 1) between the concentrations of PP vs. AER-PP and PP vs. BD-PP. Predicted AER-PP and BD-PP concentrations were calculated for all of the individual runoff samples taken from the fields during the study periods, multiplied by the volume of the runoff that the sample in question represented, and converted to annual losses by summing the P masses estimated for the individual samples.
For the annual AER-PP and BD-PP loss estimates, lower and upper predicted concentrations were calculated with a 95% prediction interval coverage and by summing these estimates in the same way as the average predicted values to lower and upper limits. The limits of prediction were calculated according to the following equation (Johnson, 1994):
 | [1] |
is the estimated average AER-PP or BD-PP concentration of an individual runoff sample, and is based on the PP vs. AER-PP and PP vs. BD-PP relationships as given in Table 2. The terms before the radical sign, t
/2, n - 2 and SEregr, refer to the tabulated t value and the standard error of estimate of the PP vs. AER-PP and PP vs. BD-PP least-squares regressions. Under the radical sign, n is the number of observations, PPobs is the measured PP concentration of the individual sample taken from the field for laboratory analyses, and PPaver is the mean PP concentration of the data sets used to calculate the PP vs. AER-PP and PP vs. BD-PP relationships. The divisor, Sxx, refers to the sum of squares of deviations from the average PP concentration of the above relationships, calculated as follows:
 | [2] |
The values of the terms included in Eq. [1] are listed in Table 2. The PP vs. AER-PP relationships were linear, so they were used as such, whereas logarithm transformation was made to linearize the PP vs. BD-PP relationships.
When the annual loss estimates were calculated, extrapolation outside the data range of Fig. 1 was done for some samples that contained more PP than predicted by the equations. At Aurajoki, there were five runoff samples with 4.9 to 22 mg L-1 PP, taken during the September 1997 to August 1998 period, that represented 5.2% of the runoff volume during that period. At Jokioinen, a single drainflow sample with 2.9 mg L-1 PP, collected during the period of September 1998 to August 1999, was not in the range of the PP vs. BD-PP relationship; this sample represented 0.6% of the period's runoff. At Lintupaju, extrapolation outside the AER-PP and BD-PP data range was done for all study years, as PP concentrations > 1.8 mg L-1 (up to 17 mg L-1) were measured for 2 to 10 samples each year. These samples represented 5.1, 1.7, 3.4, and 20% of the runoff during the four consecutive study years. At Sjökulla, all of the samples used in calculations of AER-PP and BD-PP losses had a PP concentration within the range plotted in Fig. 1.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Surface Runoff Losses of Phosphorus Forms and Their Relative Contribution to Bioavailable Phosphorus Losses
At all sites, PP loss dominated over DRP: 73 to 94% of the runoff P was associated with particulate matter. Of the PP, a fraction of 6 to 10% was desorbed in extraction by AER, whereas 34 to 58% of PP was solubilized during BD extraction. Assessment of bioavailable P losses by the sum of DRP and AER-PP would result in 4-yr averages of 669, 62, and 257 g ha-1 in surface runoff at Aurajoki, Jokioinen, and Lintupaju, respectively, and the contribution of DRP would range between 42 and 85% of bioavailable P loss. Were the sum of DRP and BD-PP used to compose the bioavailable runoff P pool, the loss estimates for bioavailable P would be more than twofold, averaging 1560, 145, and 646 g ha-1 at Aurajoki, Jokioinen, and Lintupaju, respectively. Here, DRP would typically contribute to only about one-third (14–43%) of the bioavailable P losses.
More P was lost from the Aurajoki field than from the other two sites where observations continued for 4 yr (Table 3). This was because of the greater surface runoff volumes and, as the Aurajoki field had the highest P status (Olsen P, 69–82 mg kg-1) of the fields studied, to the higher concentrations of DRP than elsewhere. Moreover, erosion was marked from the Aurajoki field, and the P content and extractability of Aurajoki runoff sediment P (Table 4) were higher than at the other sites. Runoff sediment with the lowest concentration of P was received from the Jokioinen field, which had the lowest Olsen-P status (31–45 mg kg-1) of the soils studied. Lintupaju and Sjökulla runoff sediments were very similar to each other, with P content and extractability, as well as Olsen-P level (35–45 mg kg-1), between those measured for Aurajoki and Jokioinen runoff sediments and soils.
|
Losses of Phosphorus in Surface and Subsurface Pathways
The Jokioinen data including subsurface P transport, summarized in Fig. 2
, reveal the significance of drainflow as a carrier of P from this soil. This was true for DRP as well as the different PP pools (AER-PP and BD-PP). Even though each millimeter of drainflow was assigned to a somewhat lower DRP loss (0.37 g mm-1 per hectare and year) than surface runoff (0.53 g mm-1), drainage discharge typically overwhelmed that via the surface pathway (Fig. 2) and, hence, DRP losses were greater via the drainage system. Subsurface transport of PP (and thus of AER-PP and BD-PP) was also greater than that via surface runoff in every study year (Fig. 2), but the sediment collected from the subsurface drains contained 10% less P than did the sediment transported by surface runoff (Table 4).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
From the data of Uusitalo and Ekholm (2003), who compared AER-extraction and algal assays in runoff P analysis, it can be recalculated that the amounts of PP utilized by Selenastrum capricornutum during 3-wk assays were almost 1.9-fold the AER-PP yield of 17 turbid agricultural runoff samples (largely runoff from the Aurajoki, Jokioinen, and Sjökulla fields). Were the AER-PP loss estimates of the present work recalculated to algal-available PP losses according to the results of Uusitalo and Ekholm (2003), DRP and algal-available PP would have constituted an equal share of bioavailable P losses in runoff. The short-term bioavailability of sediment-associated P is controlled by the residence time of particles in the photic zone (Williams et al., 1980), and we can speculate whether the ambient conditions (DRP concentration, light, temperature, other nutrients, pH) in natural waters are as favorable for PP utilization as they are in laboratory assays. However, considering the potential bioavailability of PP, the AER-PP losses presented in Table 3 are underestimations.
In agreement with the above statement, our estimates of the relative bioavailabilities of sediment-associated P by AER extraction (with its 6–10% share of PP) are lower than the values presented by Dorich et al. (1985) for storm runoff sediment from agricultural areas in the Black Creek watershed in Indiana (21–24% of PP being algal-available P in a 14-d assay), by Cowen et al. (1978) for PP in water samples from rivers and streams in the Lake Ontario catchment (9–45% of PP algal-available in 18 d), and by Cowen and Lee (1976) for particulate matter isolated from urban runoff (3-wk algal assays yielding mean 30% of PP, range 8–55%). Likewise, Huettl et al. (1979) found that 20 to 50% of soil P was algal-available in 24 h when the fine-earth (<20 µm) fractions of five soils were suspended in water (1 g L-1) as simulated field runoff samples. For soil suspensions (0.09–0.15 g L-1) made up of the Aurajoki and Lintupaju field soils, Krogerus and Ekholm (1999) found that 17 to 24% of P was algal-available in a 3-wk dual culture assay. Although these estimates do not deviate widely (but see Ekholm, 1998), the percentage of bioavailable P in geologic materials is not universally constant; instead, it is related to the proportions of primary apatite P, secondary inorganic P associations, and organic P (Williams et al., 1980).
As to BD-PP, the study most relevant to ours is that of Pacini and Gächter (1999), who fractionated the P of stream sediment particulate matter (concentrated by filtration) according to the modified speciation scheme of Psenner et al. (1984); this P speciation includes a BD-extraction step. The authors found that 25 to 60% of P was extractable by NH4Cl and BD (in our study BD-PP accounted for a similar 34 to 58% share of PP), but they considered the sum of BD-extractable and soluble (extractable by NH4Cl) P as the minimum bioavailable PP pool. To obtain an estimate of the maximum bioavailability of sediment PP, Pacini and Gächter (1999) added the NaOH-extractable fraction to the previous sum and ended up with a figure corresponding to 51 to 73% of PP. Similarly, about 60 to 75% estimates of PP bioavailability have been presented by Dorich et al. (1985) and James et al. (2002), who also based their assessments on NaOH extractions, as such or as a part of P fractionation. Compared with the estimates presented in the above studies, our suggestions concerning maximum PP bioavailability may thus be rather cautious, even though our BD-P yields include inorganic as well as organic P that can pass through a 0.2-µm filter [it was earlier suggested (Uusitalo and Turtola, 2003) that the organic fraction would be of minor importance, because the concentration of organic C in runoff from these fields is small]. In all cases, interpretation of the significance of the BD- and NaOH-extractable P pools in eutrophication is difficult because, as solubilization of P predominantly occurs in deep water, for most phytoplankton the accessibility of redox-labile P is highly restricted (Reynolds and Davies, 2001).
Pacini and Gächter (1999) also suggested that BD-PP would characterize particles that are deposited or precipitated at the sediment-water interface (i.e., of in-stream origin), whereas eroded topsoil would be characterized by NaOH-extractable PP. However, our results show that eroded topsoil transported by runoff also contains significant amounts of BD-PP, even exceeding the NaOH-extractable P fraction in the topsoil samples (shown in Table 1), and topsoil erosion may thus directly add to the BD-PP pool of stream sediment.
Our emphasis on PP as a source of bioavailable runoff P should not lead to the conclusion that DRP losses are of minor importance in eutrophication abatement. However, control of diffuse source DRP losses is a longer-term task than the implementation of measures to reduce erosion (e.g., reduced tillage, vegetated buffers), because DRP loss reductions can probably only be achieved by decreasing the pool of labile soil P. Initially, high soil test P levels may decline relatively rapidly when plant P uptake exceeds P inputs in fertilizers or manure (Yli-Halla et al., 2002), but even then a significant decrease in soil P status takes several years. Even though moderation of soil P status would decrease DRP losses over time, bioavailability of PP would be affected to a lesser extent. As the comparison of Aurajoki and Jokioinen DRP and sediment PP concentrations suggests, the role of bioavailable PP is relatively greater in soils with low or moderate P status. Erosion control would therefore seem to be a relatively effective way of decreasing bioavailable P loading.
Drainflow was found to be an important pathway for all forms of P at Jokioinen and Sjökulla. For subsurface losses of potentially bioavailable P at Jokioinen, PP seemed to play an even more dominant role than it did in the surface pathway, in agreement with earlier findings at the Jokioinen field (Turtola and Paajanen, 1995) and elsewhere (Simard et al., 2000). Averaged across the four study years, when subsurface drainflow comprised two-thirds of annual flow (54–78% on an annual basis) and DRP losses (ranging from 46 to 76%), as much as three-fourths of the PP losses (ranging from 65 to 82% annually) was transported by drainflow at Jokioinen. At Sjökulla, which has an old tile drainage system dating from 1950, the runoff flow was more equally distributed between surface and subsurface pathways (55 and 45%, respectively) and, presumably, Sjökulla drainflow transports somewhat less bioavailable P than does surface runoff.
At Jokioinen, renewal of the drainage system in 1991 clearly shifted the dominant runoff pathway from surface to subsurface: pre-1991 surface runoff accounted for 60 to 90% of total flow, as compared with 10 to 50% after 1991 (Turtola and Paajanen, 1995). A study conducted from 1997 to 1999 (Uusitalo et al., 2001) showed that the Jokioinen drainflow sediment derived from either topsoil or the backfill of drainage trenches; we therefore conclude that drainage improvements have changed the pathways of particles and P, but that the losses have apparently decreased to a lesser extent. In Denmark, Laubel et al. (1999) found that soil matter had migrated in a similar manner from the topsoil to subsurface drains through a sandy loam soil, and the Danish experience suggests that significant losses of bioavailable PP may not be limited to the cracking clay soils discussed in our previous and current studies.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. O. Young and R. D. Briggs Phosphorus Concentrations in Soil and Subsurface Water: A Field Study among Cropland and Riparian Buffers J. Environ. Qual., January 4, 2008; 37(1): 69 - 78. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Penn, G. L. Mullins, L. W. Zelazny, and A. N. Sharpley Estimating Dissolved Phosphorus Concentrations in Runoff from Three Physiographic Regions of Virginia Soil Sci. Soc. Am. J., September 20, 2006; 70(6): 1967 - 1974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Braskerud, K. S. Tonderski, B. Wedding, R. Bakke, A.-G. B. Blankenberg, B. Ulen, and J. Koskiaho Can Constructed Wetlands Reduce the Diffuse Phosphorus Loads to Eutrophic Water in Cold Temperate Regions? J. Environ. Qual., November 7, 2005; 34(6): 2145 - 2155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Turner, M. A. Kay, and D. T. Westermann Colloidal Phosphorus in Surface Runoff and Water Extracts from Semiarid Soils of the Western United States J. Environ. Qual., July 1, 2004; 33(4): 1464 - 1472. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | < |
