Published online 1 March 2007
Published in J Environ Qual 36:532-539 (2007)
DOI: 10.2134/jeq2006.0169
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Surface Water Quality
Freezing and Drying Effects on Potential Plant Contributions to Phosphorus in Runoff
Tiffany Roberson,
Larry G. Bundy* and
Todd W. Andraski
Dep. of Soil Science, 1525 Observatory Dr., Univ. of Wisconsin, Madison, WI, 53706-1299
* Corresponding author (lgbundy{at}wisc.edu)
Received for publication May 1, 2006.
 |
ABSTRACT
|
|---|
Phosphorus (P) in runoff from landscapes can promote eutrophication of natural waters. Soluble P released from plant material can contribute significant amounts of P to runoff particularly after plant freezing or drying. This study was conducted to evaluate P losses from alfalfa or grass after freezing or drying as potential contributors to runoff P. Alfalfa (Medicago sativa L.) and grass (principally, Agropyron repens L.) plant samples were subjected to freezing and drying treatments to determine P release. Simulated rainfall runoff and natural runoff from established alfalfa fields and a grass waterway were collected to study P contributions from plant tissue to runoff. The effects of freezing and drying on P released from plant tissue were simulated by a herbicide treatment in selected experiments. Soluble reactive P (SP) extracted from alfalfa and grass samples was markedly increased by freezing or drying. In general, SP extracted from plant samples increased in the order fresh < frozen < frozen/thawed < dried, and averaged 1, 8, 14, and 26% of total P in alfalfa, respectively. Soluble reactive P extracted from alfalfa after freezing or drying increased with increasing soil test P (r2 = 0.64 to 0.68), suggesting that excessive soil P levels increased the risk of plant P contributions to runoff losses. In simulated rainfall studies, paraquat (1,1'-dimethyl-4, 4''-bipyridinium ion) treatment of alfalfa increased P losses in runoff, and results suggested that this treatment simulated the effects of drying on plant P loss. In contrast to the simulated rainfall results, natural runoff studies over 2 yr did not show higher runoff P losses that could be attributed to P from alfalfa. Actual P losses likely depend on the timing and extent of plant freezing and drying and of precipitation events after freezing.
Abbreviations: DP, dissolved reactive P in runoff • SP, soluble reactive P in plant tissue • TP, total P in plant tissue or runoff • TSP, total soluble P in plant tissue
|
INTRODUCTION
|
|---|
PHOSPHORUS (P) in runoff from cropland is an environmental concern because it can contribute to eutrophication of natural waters and stimulate weed and algae growth in lakes and streams (Carpenter et al., 1998). The portion of plant P in the soluble fraction is one of several P sources that may contribute to P in runoff. Sharpley (1981) showed that a substantial portion of the soluble P in simulated runoff (14 to 94%) could be attributed to actively growing plant leachate P contained in agricultural crops such as cotton (Gossypium hirsutum L.), sorghum (Sorghum sudanense), and soybean (Glycine max L.). Schreiber (1999) showed that significant amounts of soluble P were leached from corn (Zea mays L.) residues left on the soil surface after grain harvest when these residues were subjected to simulated rainfall.
Several studies report increases in plant soluble P following the onset of freezing temperatures in both native and agricultural systems. White (1973) reported that freezing increased nutrient release (including P) in leachate from native vegetation if it was actively growing when frozen. Nutrient release from dried or mature vegetation after freezing was less important because cell rupture and release of soluble nutrients was limited by the lower water content of the plant cells when frozen. In South Dakota, White and Williamson (1973) showed that P in runoff from cultivated, fertilized plots cropped with oat (Avena sativa L.), corn, alfalfa (Medicago sativa L.), or fallowed was similar to runoff P content from native prairie if the prairie was subject to periodic burning. Singer and Rust (1975) found that the soluble P load in runoff from a deciduous forest in Minnesota was highest during spring snowmelt when soils were frozen. The source of the soluble P was most likely from leaf litter on the soil surface, which had been exposed to many freeze–thaw cycles resulting in cell rupture and soluble P release. The authors noted that runoff containing sufficient P to impact water quality in streams and downstream lakes occurs from forest systems.
In Wisconsin, Wendt and Corey (1980) found substantial amounts of dissolved P in simulated runoff from alfalfa after plants were frozen at the end of the growing season. Timmons et al. (1970) reported that the amount of soluble P leachate in fresh alfalfa and bluegrass (Poa pratensis L.) greatly increased following drying or freezing of tissue. Interestingly, they also observed that freezing bluegrass followed by thawing increased soluble nutrients in a subsequent leaching by about the same amount as drying the plant material. They noted that freezing and thawing followed by drying released 80% of the total plant P from bluegrass in a water-soluble form. In Nebraska, Muir et al. (1973) found a significant correlation (r = 0.45) between P concentration in stream water samples and the extent of legume production in the region represented by the surface water sample. They speculated that the higher stream P concentrations could be due to leaching of P from alfalfa during the time period when alfalfa is dormant. Gburek and Heald (1974) found that about 60 to 65% of the soluble P outputs from an agricultural watershed in Pennsylvania occurred during periods of high streamflow (late winter to early spring).
Since alfalfa occupies more than one million ha in Wisconsin (Wisconsin Department of Agriculture, Trade, and Consumer Protection, 2004), and grass pastures, roadsides, buffer strips, and waterways represent substantial additional areas, investigation of this potential source of P in runoff deserves further research attention. The objective of this study was to determine the extent of P losses from alfalfa and grasses after freezing or drying in laboratory and field experiments and the effect of management practices (topgrowth height and removal) on these losses.
|
MATERIALS AND METHODS
|
|---|
Research was conducted from 2001 to 2003 on a Plano silt loam soil (fine-silty, mixed mesic Typic Argiudolls) at the University of Wisconsin Agricultural Research Station at Arlington. This study included seven fields in established alfalfa (sites 1 through 7) and a grass waterway (site 8). Alfalfa varieties were adapted for the region and stands were 2 to 4 yr old. Site 8 consisted of mixed grasses, of which quackgrass (Agropyron repens L.) was the predominant species and smooth bromegrass (Bromus inermis Leyss.) and orchardgrass (Dactylis glomerata L.) were secondary species.
Plant Tissue Phosphorus
Above-ground tissue samples were collected from four replicates in July (sites 1, 2, 3, 4, 6, and 8) and early October (sites 5 and 7). For sites 1 through 4, four randomly selected areas in each of four established alfalfa fields were sampled. At sites 5 through 8, samples were collected from plots within a randomized complete block design. Alfalfa was in the bud to early-bloom stage of development for the July sampling and in a pre-bud vegetative stage for the October sampling. Grass was about 45 cm tall and in the pre-boot stage. Alfalfa and grass tissue samples of approximately 150 g fresh weight were subjected to several treatments before extraction with deionized (DI) water to determine soluble P content. Treatments included: (i) fresh plant extraction; (ii) extraction after freezing (24 h at –5°C); (iii) extraction after freezing and thawing (24 h at room temperature after freezing and before extraction); and (iv) extraction after drying in a forced air dryer at 60°C for 48 h. Additional treatments included field frozen alfalfa at two sampling times (site 5) and a herbicide treatment to simulate freezing (sites 6 and 7). The herbicide treatment consisted of paraquat (1,1'-dimethyl-4, 4'-bipyridinium ion) applied with a pressurized sprayer at 0.28 kg a.i. ha–1. The mode of action of paraquat is to rupture cell membranes (Weed Science Society of America, 1994), which is generally similar to the effect of freezing. Plant samples from the herbicide treatment were collected 3 d following paraquat application.
Post-treatment tissue samples were extracted with 1300 mL of DI water in 1500-mL wide mouth jars by shaking for 1 h and filtering the extracts through Whatman no. 5 filter paper. The filtered extracts were analyzed for soluble reactive P (SP) using the ascorbic acid method (Murphy and Riley, 1962) and for total soluble P (TSP) using an ammonium persulfate-sulfuric acid digestion method (USEPA, 1993). Subsamples of the fresh tissue from each site were also dried at 60°C, ground to pass a 1-mm screen, digested with nitric acid and peroxide, and analyzed for total P (TP) by inductively coupled plasma optical emission spectrometry (ICP–OES). The fresh and dry weights of these samples were also used to determine dry matter concentration in the fresh alfalfa plant tissue. All plant P measurements including TP, SP, and TSP are reported on a dry matter basis (mg P kg–1 of dry matter). Alfalfa leaves and stems were analyzed separately for SP, TSP, and TP at site 2 to determine differences in P distribution and solubility among plant parts. Soil samples (0 to 15 cm) were collected from each replicate at each site and dried at 32°C, ground to pass a 2-mm sieve, and extracted for soil test P measurements using the Bray P-1 method (Frank et al., 1998).
Simulated Runoff Phosphorus
Phosphorus losses in runoff from alfalfa and grass using simulated rainfall techniques were determined in field experiments using a randomized complete block design with four replications in July 2001. Treatments in the alfalfa experiment (site 6) included: (i) untreated alfalfa at early bloom stage (full growth); (ii) alfalfa removed by cutting at ground level (removed); (iii) alfalfa treated with paraquat 3 d before simulated rain (paraquat 3d); (iv) alfalfa treated with paraquat 3 d before simulated rain, with 12.5 mm of rainfall applied immediately before the simulated rainfall treatment (paraquat 3d+r); and (v) alfalfa treated with paraquat 10 d before simulated rain (paraquat 10d). Treatments in the grass experiment (site 8) included: (i) untreated grass (full growth); and (ii) grass treated with paraquat 3 d before simulated rain (paraquat 3d).
Steel plot frames (91 cm long by 91 cm wide by 30 cm height) were set in the soil at a 15-cm depth before simulated rain was applied. Simulated rainfall was applied using a portable, multiple intensity rainfall simulator (Meyer and Harmon, 1979) equipped with a Veejet 80150 nozzle (Spraying Systems, Wheaton, IL) located 3 m above the soil surface delivering an application rate of 75 mm h–1 with a corresponding energy of 0.278 MJ ha–1 mm–1. This rainfall intensity and duration has a recurrence interval of about 50 yr for southern Wisconsin (Huff and Angel, 1992). Runoff was collected on the down-slope side of the plot frame and continuously removed by a 0.02 MPa vacuum (Dixon and Peterson, 1968) and placed in a holding tank. Runoff was collected for a 60-min period following the onset of simulated rainfall, and the total volume of runoff from each plot was recorded. The amount of time between the onset of rainfall simulation and runoff initiation ranged from 4 to 7 min and was not significantly (p = 0.05) affected by treatments. After mixing to resuspend sediment, subsamples of the runoff were obtained for sediment, dissolved reactive P (DP), and TP determinations. The subsample for DP determination was filtered (0.45-µm pore diam.) immediately in the field. Subsamples for TP determination were acidified to 0.01 M H2SO4 (USEPA, 1993). All subsamples were frozen until analyses were performed. Sediment concentration in runoff was determined by weighing before and after drying at 105°C. Dissolved reactive P in runoff filtrate samples was determined using the ascorbic acid method (Murphy and Riley, 1962). Total P was determined by ammonium persulfate and sulfuric acid digestion on aliquots of unfiltered runoff suspension (USEPA, 1993).
Natural Runoff Phosphorus
Phosphorus losses in natural runoff from alfalfa were determined in field experiments using a randomized complete block design with four replications for a 1-yr period beginning in October 2001 (site 5) and October 2002 (site 7). Monthly air temperature and precipitation departures from the 30-yr average for 2001 to 2003 are shown in Table 1. Treatments were established by performing the following operations at the onset of the studies in October and included: (i) untreated (full growth) alfalfa (sites 5 and 7); (ii) alfalfa removed by cutting at ground level (sites 5 and 7); (iii) alfalfa cut to 20-cm height (site 5); and (iv) alfalfa treated with paraquat (site 7). Alfalfa was harvested from all treatments at the same times as production harvests (8-cm cutting height for treatments and production fields), and included three cuttings in each of the growing seasons. Alfalfa harvest dates were 14 June, 11 July, and 14 August for site 5 in 2002 and 27 May, 14 July, and 3 September for site 7 in 2003.
View this table:
[in this window]
[in a new window]
|
Table 1. Monthly temperature and precipitation departures from the 30-yr average for 2001 through 2003 at Arlington, WI.
|
|
Natural runoff collection units consisted of galvanized steel panels (1 m long by 23 cm height) inserted in the soil to a 15-cm depth (one on the upslope perpendicular to the slope and one on each side parallel to the slope). Another panel (1 m long by 15 cm height) was inserted flush with the soil surface on the down-slope edge of the 1 m2 plot area and fitted with a galvanized steel gutter (1 m long) to collect runoff. The runoff collection gutter was fitted with a 3-cm diam. outflow tube connected to vinyl tubing (3-cm i.d.) leading to a 115-L covered galvanized pail inserted flush with the soil surface containing an 8-L polyethylene collection bucket. Placement of the gutter, tubing, and collection bucket allowed runoff to flow via gravity. Coarse screening material was placed within the runoff collection gutter and finer screening material covered the collection bucket to minimize contamination from residue and insects. The collection gutter was covered with an acrylic panel to prevent rainfall and snow accumulation/melting directly into the collection unit. Runoff was collected immediately following major precipitation events from October 2001 to September 2003. Measurements included runoff volume, and sediment, DP, and TP concentrations and loads.
An analysis of variance using PROC ANOVA and PROC GLM, and regression analysis using PROC REG (SAS Institute, 1992) were performed on the data. Significant differences among treatment means were evaluated using a protected least significant difference (LSD) test or Duncan's multiple range test at the 0.05 probability level. Orthogonal contrasts were used to partition treatment effect sums of squares into single degree of freedom contrasts for cumulative DP load in natural runoff.
|
RESULTS AND DISCUSSION
|
|---|
Freezing and Drying Effects on Extractable Phosphorus in Alfalfa and Grass
Soil test P and TP, SP, and TSP in alfalfa and grass tissue for eight sites is shown in Table 2. Soil test P levels ranged from 36 to 203 g kg–1 and TP content of alfalfa and grass ranged from 1585 to 5300 mg kg–1 among sites. Soil test P values at all of the experimental sites were in the nonresponsive range for alfalfa production according to Wisconsin interpretations (Kelling et al., 1998). Fresh tissue contained 6 to 153 mg SP kg–1 (average 1% of TP) and 10 to 264 mg TSP kg–1 (average 3% of TP). Significant increases in SP and TSP content due to freezing, freezing and thawing, and/or drying of tissue occurred at all sites and treatment effects were similar for alfalfa and grass (Table 2). Frozen tissue contained 100 to 839 mg SP kg–1 (average 8% of TP) and 105 to 1281 mg TSP kg–1 (average 14% of TP). Frozen/thawed tissue contained 167 to 1047 mg SP kg–1 (average 14% of TP) and 167 to 1793 mg TSP kg–1 (average 18% of TP). Dried tissue contained 494 to 1550 mg SP kg–1 (average 26% of TP) and 652 to 2873 mg TSP kg–1 (average 43% of TP). For most sites, tissue treatment effects on SP and TSP concentrations followed the order: fresh < frozen 
frozen/thawed < dried.
View this table:
[in this window]
[in a new window]
|
Table 2. Tissue treatment effects on soluble reactive P and total soluble P concentrations in alfalfa (sites 1 through 7) and grass (site 8) samples at eight sites with a range of soil test P and tissue total P levels obtained in July (sites 1, 2, 3, 4, 6, and 8) and October (sites 5 and 7).
|
|
In previous work, Timmons et al. (1970) reported that freezing bluegrass followed by thawing greatly increased SP that could be extracted relative to fresh vegetation and released about the same amount of SP as drying the plant material. We did not include a frozen/thawed grass treatment in our study; however; our results indicate that drying alfalfa tissue results in significantly higher SP and TSP compared with freezing/thawing alfalfa tissue. Sharpley (1981) found that 7 to 9% of the P leached from fresh plants by simulated rainfall was soluble organic P. In our study, soluble organic P (TSP–SP) in fresh alfalfa tissue was much higher likely due to a more vigorous extraction method. Soluble organic P was 9% of TSP in grass and ranged from 18 to 44% (average 33%) of TSP in alfalfa.
Soluble reactive P content in field frozen alfalfa (site 5) was significantly higher in samples collected within several days following a killing frost in October (338 mg kg–1) compared with samples collected from the same site in December (159 mg kg–1) (Table 2). There were 18 d with low temperatures below 0°C between the 7 Oct. and 10 Dec. 2001 alfalfa sampling dates for site 5. Precipitation likely resulted in some leaching of SP from alfalfa tissue during the period between sampling times. The application of paraquat to alfalfa and grass resulted in similar SP and TSP contents as the dried tissue treatment, which was significantly higher than the frozen tissue treatment (Table 2). These results indicated that paraquat treatment provided a better indication of SP release from plant tissue by drying than by freezing.
The effect of freezing and drying on SP and TSP in alfalfa leaves and stems was determined for site 2 (Fig. 1). Total P concentration was 3450 mg kg–1 in leaves and 3725 mg kg–1 in stems. The percentage of TP as SP was significantly higher in fresh leaves (4%) compared with stems (1%), but was similar in leaves and stems for frozen, frozen/thawed, and dried tissue. The percentage of TP as TSP was significantly higher in fresh leaves (6%) compared with stems (1%) and in dried leaves (96%) compared with stems (55%), but similar for frozen and frozen/thawed tissue. The treatment average soluble organic P fraction of TSP was 37% for leaves and 23% for stems. The distribution of dry matter between leaves and stems is approximately equal in a three harvest per year alfalfa system harvested at the early flower stage (Sheaffer et al., 2000). Therefore, our results suggest that the alfalfa leaves have a slightly greater potential for leachate P losses compared with stems.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 1. Percentage of total P (TP) concentration as soluble reactive P (SP) and total soluble P (TSP) in fresh (F), frozen (Fr), frozen and thawed (Fr/T), and dried (D) alfalfa leaves and stems from site 2. Total P concentration was 3450 mg kg–1 in leaves and 3725 mg kg–1 in stems (p = 0.18). Error bars represent the standard deviation of four replicates.
|
|
The relationship between soil test P and TP, SP, and TSP in fresh, frozen, frozen and thawed, and dried alfalfa tissue from six sites (1, 2, 3, 4, 6, and 7) is shown in Table 3. Relationships between soil test P and various soluble P measurements in alfalfa are likely influenced by the proportion of TP extracted by the procedure. A much higher percentage of plant total P is released when plant tissue is frozen or dried relative to fresh tissue where only a small percentage (1 to 3% of TP) is released as soluble P. The relationship with soil test P likely improves when more of the total plant P is extracted as is the case following freezing or drying.
View this table:
[in this window]
[in a new window]
|
Table 3. Relationship between soil test P (0 to 15 cm) and total, soluble reactive, and total soluble P concentrations in alfalfa using fresh, frozen, frozen and thawed, and dried tissue at six sites (n = 24).
|
|
A significant but weak relationship between soil test P and alfalfa TP content was observed (r2 = 0.39). Soil test P was strongly related to SP and TSP concentrations in frozen or dried tissue (r2 = 0.64 to 0.68) and intermediate in frozen/thawed tissue (r2 = 0.46 to 0.48). Soil test P level was not related to SP or TSP in fresh tissue (r2 = 0.01 to 0.02). The relationship between soil test P and SP in frozen alfalfa tissue is shown graphically in Fig. 2.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 2. Relationship between soil test P (0 to 15 cm) and soluble reactive P (SP) concentration in frozen alfalfa tissue from six sites.
|
|
Previous work has either measured increased P losses from vegetation after freezing (Timmons et al., 1970; White, 1973; Singer and Rust, 1975; Wendt and Corey, 1980) or suggested that this may be a potential source of P in runoff (Muir et al., 1973). Results from our study confirm that SP released from plants by freezing or drying is a potential source of P in runoff from cropland. It is a widely held view that alfalfa does not accumulate P beyond the level needed to maximize dry matter production. However, results from this study show that increasing soil test P levels increased the amount of SP released when plants are frozen or dried. Lowering soil test P or avoiding excess soil P accumulation could help minimize the potential contribution of plant P to P losses in runoff.
Simulated Runoff Phosphorus Losses in Alfalfa and Grass
Runoff amount and sediment, DP, and TP concentrations and loads in runoff following 60 min of simulated rainfall for several treatments in alfalfa (site 6) and grass (site 8) are shown in Table 4. Runoff amounts were significantly higher in the removed and paraquat (10d) treatments compared with the full growth and paraquat (3d) treatments. The greater runoff was likely due to less surface cover (removed) and lower leaf surface area resulting from senescence (paraquat 10d). The combination of these factors could have increased raindrop energy and enhanced soil surface sealing. Sediment concentrations and loads were significantly higher in the paraquat treatments. The higher sediment losses are likely due to increased tissue organic matter in runoff since sediment losses were lower in the removed treatment where higher soil losses would be expected. Removal of alfalfa topgrowth (removed) did not significantly increase DP concentrations in runoff compared with the control (full growth), but did increase TP concentrations. Removing alfalfa increased DP and TP loads in runoff, likely due to greater runoff volume and extraction of soil P. The intent of the alfalfa removed treatments was to remove or reduce plant P as a contributor to runoff P. Our results indicate that the reduction in plant P was offset by increased soil P runoff contributions. Paraquat-treated alfalfa significantly increased DP and TP concentrations and loads to levels similar to, or greater than, the removed treatment. Increasing the time between paraquat treatment and simulated rainfall or including an additional rainfall event had no effect on DP in runoff, but TP losses were higher at 10 d compared with 3 d due to greater runoff. Runoff amounts were much greater in alfalfa (9 to 32 mm) compared with grass (1 to 2 mm). Paraquat treatment of grass increased DP and TP concentrations in runoff, but did not significantly influence other runoff measurements. Runoff P losses were similar in full growth alfalfa and grass. Results from the rainfall simulator and laboratory portions of this study support the hypothesis that freezing or drying of alfalfa tissue increases the potential for increased SP losses in runoff from cropland. The use of paraquat appears to simulate drying, rather than freezing, effects on SP release from tissue.
View this table:
[in this window]
[in a new window]
|
Table 4. Alfalfa (site 6) and grass (site 8) treatment effects on runoff amount and sediment and dissolved reactive P (DP) and total P (TP) concentrations and loads in simulated runoff, July 2001.
|
|
Rainfall simulation studies in Wisconsin tilled corn production systems report that DP generally comprises <15% of TP, and that TP is strongly correlated with sediment, but not with DP (Bundy et al., 2001; Andraski and Bundy, 2003; Andraski et al., 2003). In this study, DP comprised 50% of TP load in runoff under full growth alfalfa as a result of low TP concentrations and loads in runoff. The low TP losses in alfalfa are primarily due to minimal sediment/particulate P losses compared with tilled corn production systems where TP loads are usually substantially higher than in alfalfa production systems.
Natural Runoff Phosphorus Losses in Alfalfa
Annual natural runoff amounts ranged from 27 to 44 mm among treatments and years (Tables 5 and 6), representing 4 to 7% of annual precipitation (Table 1). Alfalfa treatment effects on cumulative runoff, sediment, DP, and TP loads were not significant for site 5 (Oct. 2001 to Sep. 2002) or site 7 (Oct. 2002 to Sep. 2003) (Tables 5 and 6). However, it is interesting to note the similar relative effects of the full growth and removed treatments on DP, TP, and runoff loads in simulated and natural runoff (Table 4 and Fig. 3 and 4).
View this table:
[in this window]
[in a new window]
|
Table 5. Effect of fall cutting height of alfalfa on cumulative runoff, sediment, dissolved reactive P (DP), and total P (TP) loads in natural runoff at site 5 from 17 collection dates from Oct. 2001 through Sept. 2002.
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. Effect of fall cutting height and paraquat application to alfalfa on cumulative runoff, sediment, dissolved reactive P (DP), and total P (TP) loads in natural runoff at site 7 from 14 collection dates from Oct. 2002 through Sept. 2003.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3. Effect of fall cutting height of alfalfa (5 Oct. 2001) on cumulative dissolved reactive P (DP) (solid line) and natural runoff (dashed line) load at site 5, Oct. 2001 through Sept. 2002. (*, **, ***) Orthogonal comparison of full growth and removed treatment on cumulative DP load was significant at the 0.05, 0.10, and 0.20 probability levels, respectively.
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 4. Effect of fall cutting height and paraquat application to alfalfa (15 Oct. 2002) on cumulative dissolved reactive P (DP) (solid line) and natural runoff (dashed line) load at site 7, Oct. 2002 through Sept. 2003. (*) Orthogonal comparison of full growth and paraquat treatment on cumulative DP load was significant at the 0.20 probability level.
|
|
At site 5, snowmelt within a 1-mo period (25 Jan. to 20 Feb.) was the major contributor to the annual cumulative DP load (average of 53%) during which time 75% of the annual cumulative amount of runoff occurred (Fig. 3). An average of 30% of the annual cumulative DP load occurred during the mid-summer period (26 Jul. to 3 Sep.) even though runoff represented only about 1% of the annual cumulative amount of runoff. Higher DP concentrations during this period may have been due to a greater influence of soil P where runoff events occurred shortly after alfalfa harvests (11 Jul. and 14 Aug.). Similarly at site 7, snowmelt from two runoff events (20 Feb. and 20 Mar.) contributed 68% of the annual cumulative DP load where 50% of the annual cumulative amount of runoff occurred (Fig. 4). Marked DP losses also occurred on 16 July due to high runoff amounts (24% of the cumulative).
Significant treatment effects on cumulative DP load in natural runoff for any given sampling date (Fig. 3 and 4) could not be detected due to the inherent variability in the amount of natural runoff as indicated by the high coefficients of variability (40 and 71%) (Tables 5 and 6).
Independent treatment comparisons were performed to more critically evaluate treatment effects on natural runoff measurements. Orthogonal contrast determinations indicated the highest probability (p = 0.05 to 0.20) of cumulative DP load differences occurred when comparing removed vs. full growth alfalfa following several runoff events occurring from 10 Dec. 2001 to 6 June 2002 at site 5 (Fig. 3). The highest probability (p = 0.20) of cumulative DP load differences at site 7 occurred when comparing paraquat and full growth alfalfa following two runoff events from 20 Mar. to 11 Apr. 2003 (Fig. 4).
A correlation matrix of natural runoff measurements for sites 5 (n = 159) and 7 (n = 180) is shown in Table 7. Sediment and P loads were strongly correlated with the amount of runoff, but not with sediment or P concentrations with the exception of sediment at site 7. A strong correlation between DP and TP concentration and load occurred due to the relatively low sediment losses resulting in a high percentage of TP as DP (44 to 62%) in natural runoff.
View this table:
[in this window]
[in a new window]
|
Table 7. Correlation matrix for concentrations and loads of dissolved reactive P (DP), total P (TP), sediment, and runoff amounts for natural runoff events in alfalfa for sites 5 and 7, 2001 through 2003.
|
|
The absence of strong natural P runoff response to various alfalfa treatments seems to conflict with the findings of Wendt and Corey (1980). However, the difference in measurement techniques between simulated rainfall used by Wendt and Corey (1980) and the natural runoff collection method used in this study could account for the difference in findings. Climatic conditions, including the timing and extent of plants freezing and drying, and precipitation events after freezing, likely influence the potential for P losses in natural runoff.
Since our laboratory and simulated rainfall runoff work demonstrate the potential for plant P losses in runoff, climatic conditions involving freezing conditions followed by precipitation events large enough to produce runoff would likely result in significant losses of P from plant materials. In our natural runoff studies, alfalfa topgrowth was frozen before 1 November in 2001 and 2002, but the absence of significant runoff events following plant freezing likely minimized P losses. As shown in Fig. 3 and 4 and discussed above, the larger runoff amounts in our natural runoff experiments occurred in conjunction with snowmelt events during the January to March time period. Moderate precipitation events that occur after plant freezing but before the soil is frozen may not generate runoff and would likely lower the risk of runoff P losses by moving plant P released by freezing into the soil where it would be less susceptible to runoff loss. This is in contrast to conditions created in our simulated rainfall work where high intensity rainfall created a runoff event resulting in losses of plant-derived P in the runoff.
|
CONCLUSIONS
|
|---|
Our results confirm that SP released from plants by freezing or drying is a potential source of P in runoff from cropland. Freezing, freeze–thaw, and drying treatments imposed in this study greatly increased water SP extracted from alfalfa, and more limited work suggests that P losses from grasses after freezing or drying are as great as from alfalfa. In regions with substantial areas occupied by these plant species and with climatic conditions resulting in freezing and/or drying of the plant material, considerable potential exists for runoff P contributions from this source. We found that SP extracted from alfalfa after freezing or drying increased with increasing soil test P indicating that the potential for alfalfa contributions to runoff P were higher at excessive soil test P levels. Simulated rainfall runoff studies using the herbicide paraquat to simulate freezing and/or drying of alfalfa and grasses showed that this treatment increased DP and TP concentrations in runoff. With alfalfa, runoff DP load was also increased by the paraquat treatment, and TP loads were higher than or equal to those from the untreated control. Paraquat appears to better simulate the effects of drying rather than of freezing on SP release from plants. The contribution of DP to runoff TP loads in this study with alfalfa was much higher (50 vs. 15%) than the DP contribution to TP loads in previously conducted work in corn production systems. In contrast to the simulated rainfall results, natural runoff collected over a continuous 2-yr period, showed no significant increase in runoff P losses due to alfalfa plant P contributions. While the potential for plant P losses in runoff exists, climatic factors, including the timing and extent of plant freezing and drying and of precipitation events after freezing, are likely key determinants of actual P losses.
|
NOTES
|
|---|
Research supported by the Wisconsin Dep. of Agric., Trade, and Consumer Protection, the Univ. of Wisconsin Nonpoint Pollution and Demonstration Project, and the College of Agric. and Life Sci., Univ. of Wisconsin-Madison. The senior author gratefully acknowledges receipt of an Advanced Opportunity Fellowship from the Univ. of Wisconsin Graduate School. The authors gratefully acknowledge J.S. Studnicka for technical support and contributions from the staff at the University of Wisconsin Agricultural Research Station at Arlington.
|
REFERENCES
|
|---|
- Andraski, T.W., and L.G. Bundy. 2003. Relationships between phosphorus levels in soil and in runoff from corn production systems. J. Environ. Qual. 32:310–316.[Abstract/Free Full Text]
- Andraski, T.W., L.G. Bundy, and K.C. Kilian. 2003. Manure history and long-term tillage effects on soil properties and phosphorus losses in runoff. J. Environ. Qual. 32:1782–1789.[Abstract/Free Full Text]
- Bundy, L.G., T.W. Andraski, and J.M. Powell. 2001. Management practice effects on phosphorus losses in runoff in corn production systems. J. Environ. Qual. 30:1822–1828.[Abstract/Free Full Text]
- Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W. Howarth, A.N. Sharpley, and V.H. Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecol. Appl. 8:559–568.
- Dixon, R.M., and A.E. Peterson. 1968. A vacuum system for accumulating runoff from infiltrometers. Soil Sci. Soc. Am. Proc. 32:123–125.
- Frank, K., D. Beegle, and J. Denning. 1998. Phosphorus. p. 21–26. In J.R. Brown (ed.) Recommended chemical soil test procedures for the North Central Region. NC Regional Res. Publ. no. 221 (Rev.).
- Gburek, W.J., and W.R. Heald. 1974. Soluble phosphate output of an agricultural watershed in Pennsylvania. Water Resour. Res. 10:113–118.
- Huff, F.A., and J.R. Angel. 1992. Rainfall frequency atlas of the midwest. Bull. 71. Illinois State Water Survey, Champaign.
- Kelling, K.A., L.G. Bundy, S.M. Combs, and J.B. Peters. 1998. Soil test recommendations for field, vegetable, and fruit crops. Publ. A2809. Univ. of Wisconsin Extension, Madison.
- Meyer, L.D., and W.C. Harmon. 1979. Multiple-intensity rainfall simulator for erosion research on row sideslopes. Trans. ASAE 22:100–103.
- Muir, J., C.S. Edwin, and R.A. Olson. 1973. A study of factors influencing the nitrogen and phosphorus contents of Nebraska waters. J. Environ. Qual. 2:466–470.
- Murphy, J., and J.R. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chem. 27:31–36.[Medline]
- SAS Institute. 1992. SAS/STAT user's guide. Release 6.03/4th ed. SAS Inst., Cary, NC.
- Schreiber, J.D. 1999. Nutrient leaching from corn residues under simulated rainfall. J. Environ. Qual. 28:1864–1870.
- Sharpley, A.N. 1981. The contribution of phosphorus leached from crop canopy to losses in surface runoff. J. Environ. Qual. 10:160–165.
- Sheaffer, C.C., N.P. Martin, J.F.S. Lamb, G.R. Cuomo, J.G. Jewett, and S.R. Quering. 2000. Leaf and stem properties of alfalfa entries. Agron. J. 92:733–739.[Abstract/Free Full Text]
- Singer, M.J., and R.H. Rust. 1975. Phosphorus in surface runoff from a deciduous forest. J. Environ. Qual. 4:307–311.
- Timmons, D.R., R.F. Holt, and J.J. Latterell. 1970. Leaching of crop residues as a source of nutrients in surface runoff water. Water Resour. Res. 6:1367–1375.
- USEPA. 1993. Methods for the determination of inorganic substances in environmental samples. EPA/600/R-93/100. Method 365.1. USEPA, Washington, DC.
- Weed Science Society of America. 1994. Paraquat. p. 226–228. In WSSA Herbicide Handbook, 7th ed. WSSA, Lawrence, KS.
- Wendt, R.C., and R.B. Corey. 1980. Phosphorus variations in surface runoff from agricultural lands as a function of land use. J. Environ. Qual. 9:130–136.
- White, E.M. 1973. Water-leachable nutrients from frozen or dried prairie vegetation. J. Environ. Qual. 2:104–107.
- White, E.M., and E.J. Williamson. 1973. Plant nutrient concentrations in runoff from fertilized cultivated erosion plots and prairie in eastern South Dakota. J. Environ. Qual. 2:453–458.
- Wisconsin Department of Agriculture, Trade, and Consumer Protection. 2004. Wisconsin agricultural statistics. Wisconsin Agricultural Statistics Service, Madison, WI.
This article has been cited by other articles:
|
|
|
|
 
P. Fiener and K. Auerswald
Effects of Hydrodynamically Rough Grassed Waterways on Dissolved Reactive Phosphorus Loads Coming from Agricultural Watersheds
J. Environ. Qual.,
February 6, 2009;
38(2):
548 - 559.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
