| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a School of Natural Resources, The Ohio State University, 2021 Coffey Rd., Columbus, OH 43210-1085
b Environmental Soil Science, Univ. of Missouri-Columbia, 302 Anheuser-Busch Natural Resources Building, Columbia, MO 65211
c USDA-ARS, 268 Agricultural Engineering Building, Columbia, MO 65211
d 251 Agricultural Engineering Building, Columbia, MO 65211
* Corresponding author (blanco.16{at}osu.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: B-Fescue-FS, switchgrass barrier in combination with the fescue-FS • B-Native-FS, switchgrass barrier in combination with a native grass and forbs species filter strip • CCF, continuous cultivated fallow • FS, vegetative filter strips • NPS, nonpoint source
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Grass barriers show promise as an economical alternative to existing conservation practices for reducing NPS pollutants. Grass barriers are narrow strips (approximately 1.2 m) of tall, erect, stiff-stemmed, native warm-season perennial grasses planted on the field contour (Kemper et al., 1992). Barriers form natural terraces (Dabney et al., 1999), slow runoff and promote infiltration (Meyer et al., 1995), enhance deposition of soil and organic matter (Melville and Morgan, 2001), promote degradation of sediment-bound chemicals (Groffman et al., 1991), and enhance wildlife habitat (Schultz et al., 1995). This approach for reducing NPS pollution can be a less-costly alternative to terraces where slopes are not too steep.
Grass barriers differ from FS because FS are typically much wider (>5 m). Vegetative filter strips are established between field borders and waterways. Narrow-row stiff-stemmed barriers may be more acceptable to farmers because they occupy much less land than FS. In addition, short statured plants such as fescue provide little benefit to wildlife. Vegetative filter strips of native perennial, usually tall, warm-season grass species when used with barriers may afford adequate control of NPS pollutants and provide habitat for upland wildlife.
Studies on value of FS for reducing sediment, N, and P in runoff have recently been published (Dillaha et al., 1989; Daniels and Gilliam, 1996; Srivastava et al., 1996; Melville and Morgan, 2001). Laboratory (Dabney et al., 1995; Ghadiri et al., 2001) and field (Chaubey et al., 1995; Rankins et al., 2001) studies indicate that FS significantly reduce sediment and nutrient loss in runoff. The most widely used grass for FS in the USA is fescue with extensive use in midwestern states (Sleper and Buckner, 1995; Rankins et al., 2001).
Information on grass barriers is limited (Meyer et al., 1995; Dabney et al., 1999). Most studies have been conducted in the laboratory (Dabney et al., 1995; Meyer et al., 1995; Ghadiri et al., 2001). Field studies are few (McGregor et al., 1999; Eghball et al., 2000; Gilley et al., 2000). Moreover, few studies have evaluated comparative effectiveness of fescue FS vs. barriers when used in combination with fescue or native plant species FS for reducing NPS pollution (Lee et al., 2003).
Modeling NPS pollution transport through FS and grass barriers is needed to understand and predict pollution transport. While equations have been developed to estimate effectiveness of FS for trapping sediments (Tollner et al., 1977; Foster, 1982; Flanagan et al., 1989), validation of equations with field and plot data is scarce. Moreover, prediction of barrier performance for trapping sediment has received little attention.
Objectives of this study were: (i) to determine effectiveness of a fescue-FS vs. B-fescue-FS or B-Native-FS in reducing runoff, sediment, N, and P loss from 8-m long runoff plots on an Aeric Vertic Epiaqualf; and (ii) to evaluate methods to predict transport of sediment, N, and P through switchgrass barriers and fescue filter strips.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Twelve 1.5- by 16-m plots with four treatments replicated three times were arranged in a randomized complete block design (Fig. 1) . The length of the plots was oriented up- and downslope. Soil berms 200 mm in height and 250 mm in width were constructed as plot borders. Berms were treated with polyacrylamide at a rate of 9 kg ha–1, and covered with geotextile fabric to reduce berm erosion to nondetectable levels. Plots were planned with a 1.5 by 8 m sediment source-area (or source-area) managed under CCF, above a downslope area under FS or CCF of the same size (Fig. 1). Each pair of parallel plots included a 3-m alley oriented up- and downslope between plots to facilitate positioning a rainfall simulator. Glyphosate herbicide (N-phosphonomethyl-glycine) was applied at 8 L ha–1 to kill vegetation in the sediment source-area in June 2001. The source area was tilled with a rototiller to a depth of approximately 80 mm in July 2001. The source area was managed under CCF and rototilled after major rainfall events. The area below the source area in the CCF treatment was managed the same as the source area. Four treatments were (i) a check managed in CCF without switchgrass barrier or filter strip, (ii) Fescue-FS, (iii) B-Fescue-FS, and (iv) B-Native-FS (Fig. 1). The word barrier is used to reference switchgrass barriers throughout this paper.
|
Rainfall Simulation
Simulated rainfall was used to evaluate treatments in August 2002. A rotating-boom rainfall simulator was used to apply rain at an intensity of 66 mm h–1 for 1 h (Swanson, 1965). The rainfall simulator was positioned to apply rainfall to a pair of plots concurrently. The simulator uses 10 booms with 30 nozzles, with nozzles positioned at radii of 1.5, 3.0, 4.5, 6.0, and 7.6 m with 2, 4, 6, 8, and 10 nozzles, respectively, on each successive radius. Nozzles were 2.7 m aboveground and rotated in a circle while continuously spraying. Diameter of the rainfall simulator is approximately 16 m, allowing rainfall to cover both the source and filter strip areas. Rain gauges were set at 1-m intervals along the simulator boom radius to monitor rainfall distribution.
Mean simulated rainfall application was 66 ± 1.6 mm h–1. Rainfall intensity at the edge was less than near the center of the simulator (61 ± 1.2 vs. 70 ± 0.4 mm h–1). Mean wind velocity for all runs was 15 ± 2 km h–1. However, rainfall intensity near the outside edge of the application area was probably reduced by wind combined with centrifugal effects. The simulator was adjusted during the study to produce a uniform intensity throughout the area of application; however this effort was not perfect. Variable rainfall intensity with distance from the center occurred for all runs. Thus, this variability should not be a factor in assessing relative differences among treatments.
Water supplied to the simulator from a nearby lake had an electrical conductivity of 1 dS m–1. Simulated rainfall protocol began with a dry soil run at 66 mm h–1 for 1 h. A subsequent wet-run simulation was done approximately 24 h later at 66 mm h–1 for 1 h. Dry and wet runs were designed to simulate large natural rainfall events with a recurrence interval of a 10-yr return period for mid-Missouri. Granular fertilizer (13% N, 44% P, and 83% K) was applied to the source area 24 h before rainfall simulation at 80 kg ha–1 of N, 35 kg ha–1 of P, and 66 kg ha–1 K. Although no crop was grown, fertilizer application facilitated evaluation of the effectiveness of barriers and filter strips to reduce nutrient transport. This fertilizer rate is the amount that would have been applied based on the soil test if a crop had to be grown. Fertilizer was uniformly broadcast and incorporated to a depth of approximately 80 mm with a rototiller.
Runoff Collection and Sampling
Watertight runoff collectors of a V-shape (0.08 m wide, 1.5 m long, and 0.06 m deep) were constructed with angle iron for sampling runoff. Each collector was covered with a hinged metal cover fitted with a rubber gasket. The cover was fit with clamps to secure it to the trough between sampling periods. Hinges allowed the collector cover to be quickly opened and closed for runoff sampling. Collectors were affixed with four 250-mm long spikes to anchor them into the soil. Collectors were set to a 3% slope to allow runoff into collection pits containers. In the cover-closed position, runoff passed over the cover. Runoff collectors were installed across the plots at 1 m above the downslope edge of the source area and in the FS area at 0.7, 4, and 8 m below the source area. Collection pits were created to position a 4-L container for collecting runoff.
Runoff collection was done only during the wet-runs. Runoff was sampled every 10 min for 5 s at all collectors during runoff. Samples at a given time were collected first from the collector at the downslope end of the plot and then successively from each collectors upslope. This allowed sampling without affecting runoff downstream. Six samples were collected from each location for a total of 24 samples from each plot-event, totaling 432 samples from the 18 plots. Volume and weight of runoff of each sample were recorded. Runoff volume was regressed vs. time of collection, and the resulting regression equations were integrated over 0 to 60 min to compute the runoff volume on a 1-h basis. Runoff depth was computed by dividing runoff volume by the corresponding contributing area above each sampling position in accord with Sheridan et al. (1999) and Lee et al. (2003). Runoff ponding above grass treatments was measured vertically by inserting a meter stick into the pond.
Sediment, Nitrogen, and Phosphorus Analysis
Runoff samples were stirred to suspend sediments, and two aliquots were taken for analysis. A 0.5-L aliquot was used for determination of sediment concentration. Sediment concentration was measured using the evaporation method that consisted of decanting water after 48 h and drying at 105°C (Brakensiek et al., 1979). A 0.25-L aliquot of a composite of samples for each sampling position was used to determine N and P concentration. These samples were stored in an insulated cooler and transported to the laboratory within approximately 4 h after a run. Samples were filtered through Whatman #1 filter paper for determination of nitrate (NO3–N), ammonium (NH4–N), and orthophosphate (PO4–P) and then stored at 4°C until analyzed. Total N and P concentrations were determined from unfiltered aliquots. Analysis of N and P was done using a Lachat flow injection analyzer (Lachat QuikChem 800 Zellweger Analytics, Milwaukee, WI). Sediment mass and nutrients were computed as the product of runoff and concentration. Organic N was calculated as the difference of NO3–N and NH4–N from total N. Particulate P was calculated as the difference of total P minus PO4–P. Sediment loss per unit area was computed by dividing sediment by the corresponding contributing area above each sampling position.
Sediment Transport Prediction with Barriers and Fescue Filter Strips
Equations compiled by Haan et al. (1994) based on work of Tollner et al. (1977) and Flanagan et al. (1989) were used to predict sediment trapping efficiency (Ts) of Fescue-FS and B-Fescue-FS assuming: (i) steady runoff and sediment flow, and (ii) sediment deposition beginning at the upper portion of the grass strips. Inputs included incoming sediment discharge, runoff rate, density and height of vegetation, calibrated Manning's roughness, width of Fescue-FS and B-Fescue-FS (m), and soil slope. Prediction of sediment trapping considered two zones. Zone 1 was the upper portion of the strip where most sediment deposition occurs. Zone 2 was the remaining lower strip where fine sediment settles. Prediction was conducted using data collected every 10 min during the 1-h run. Eighteen data values for each grass treatment were used for the prediction, corresponding to samples collected at 1 m above, 0.7 and 4 m below the source area.
Flow depth and hydraulic radius for each zone of Fescue-FS and B-Fescue-FS (0.7, 4, and 8 m) were estimated using a calibrated Manning's equation (Tollner et al., 1977) as follows:
 | [1] |
 | [2] |
Sediment transport rate of silt and clay (qsi) in Zone 1 was calculated as
 | [3] |
The fraction of soil trapped by settling (Ts) in Zone 2 based on Reynold's Number (Re), and the fall number of soil particles (Nf in m s–1) was computed as:
 | [4] |
 | [5] |
 | [6] |
An adjusted trapping efficiency (fd) for sand, silt, and clay was estimated accounting for sediment trapped by infiltration using an effective saturated hydraulic conductivity (Keff; 0.34 mm h–1) for the Mexico claypan soil (Blanco-Canqui et al., 2002).
 | [7] |
The total trapping efficiency (fto) was computed as:
 | [8] |
Data Analysis
Statistics were calculated for runoff, sediment, and nutrient data. The General Linear Models (GLM) procedure of SAS (SAS Institute Inc., 1999) was used to test hypotheses that runoff, sediment, and nutrient reduction differences between adjacent sampling positions are the same among treatments. Orthogonal contrasts at the same sampling positions were used to compare main effects for Fescue-FS vs. B-Fescue-FS, Fescue-FS vs. B-Native-FS, and CCF vs. the mean of all grass treatments. Regression was used to study relationships of runoff, sediment, and nutrient reduction with distance. Analysis of confidence intervals of means was conducted to test differences among treatments with distance (Snedecor and Cochran, 1989).
A small decline in runoff depth with distance occurred in the downslope area of all the treatments as discussed earlier (Table 1). This difference is not expected to significantly influence treatment effects because relative amounts of runoff, sediment, and nutrients between the CCF and the three other treatments (Fescue-FS, B-Fescue-FS, and B-Native-FS) at all sampling positions were large. The CCF produced the most runoff, sediment, and nutrient loss (Tables 1 and 2). Because of the decline in runoff with distance, relative amounts of runoff, sediment, and nutrients vs. distance of Fescue-FS, B-Fescue-FS, and B-Native-FS were illustrated by using data normalized with the CCF data, allowing for comparison of reductions due only to grass treatments effects. Relative runoff, sediment, and nutrients were computed by dividing the amount leaving each sampling position (0.7, 4, and 8 m) by the amount collected at 1 m above the downslope end of the 8-m source area. Graphs were prepared using these relative values scaled vs. the CCF treatment that had no barrier or filter strip. Relative values of runoff, sediment, and nutrients were discussed as percentages.
|
|
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Runoff did not differ between the Fescue-FS vs. B-Native-FS and Fescue-FS vs. B-Fescue-FS (Table 1 and 2; Fig. 2a) at 4 and 8 m. This shows that B-Native-FS was as effective as Fescue-FS in reducing runoff. A related study by Tufekcioglu et al. (1999) reported that native species form a network of roots to a depth of 1.5 m, increasing potential for infiltration. Early stages of infiltration may be improved; however, long-term infiltration will be dominated by the very slowly permeable Bt horizon (Blanco-Canqui et al., 2002). Results suggest that switchgrass barriers in combination with Fescue-FS are a potential alternative to Fescue-FS alone for improving infiltration in claypan soils, thereby reducing surface runoff. Runoff reduction due to infiltration can have practical implications for removing fine sediments and soluble nutrients from runoff.
|
Reduction of transported sediment in the B-Fescue-FS treatment was mostly due to ponding upslope of barriers that reduced runoff velocity and promoted sediment deposition. The maximum ponding depth was 0.03 ± 0.01 m and extended 0.7 ± 0.05 m above the barriers. Runoff through B-Fescue-FS was delayed by increased detention storage created by barriers. These observations agree with Ghadiri et al. (2001) who reported that runoff ponding caused sediment deposition upslope of barriers, thereby reducing sediment loss. Runoff ponding above Fescue-FS was negligible. Fescue residues and submerged plant parts within the Fescue-FS treatment enhanced the performance of fescue for reducing sediment in runoff (Jin et al., 2002).
Sediment loss was not different between the Fescue-FS vs. B-Fescue-FS treatments at either the 4- and 8-m positions (Table 2). The 4-m strip reduced sediment loss by approximately 93% in both treatments. The 8-m strip reduced sediment loss by 97% agreeing with Coyne et al. (1995) who found that a 9-m Fescue-FS reduced 99% of sediment loss on a Maury silt loam. Performance of 4-m Fescue-FS in our study is slightly higher than that reported by Dillaha et al. (1989) who found that a 4.6-m filter strip retained only 83% of sediment. This small difference may be attributed to a much greater simulated rainfall application rate. Large rainfall events would be expected to diminish the benefit of Fescue-FS.
Sediment reduction by the 0.7-m B-Fescue-FS was equivalent to reduction by 4 m of Fescue-FS. This indicates that only 15% of the land required for Fescue-FS is needed for B-Fescue-FS for the same effectiveness, reducing the amount of land taken out of production. We conclude that narrow switchgrass barriers in combination with FS improve the performance of Fescue-FS.
Sediment loss did not differ between the Fescue-FS vs. B-Native-FS (Table 2). Both Fescue-FS and B-Native-FS reduced sediment loss by 91% within 4 m (Table 1). The width increase from 4 to 8 m reduced the sediment loss by an additional 5%. This indicates that native species filter strips, when used in conjunction with barriers, can be as effective as Fescue-FS for trapping sediment in runoff. Our results support Rankins et al. (2001) who found that effectiveness of native species and Fescue-FS for sediment reduction was the same.
Foster (1982) stated that sediment transport through grass strips diminishes exponentially with increasing grass width. The relative sediment loss shown in Fig. 2b did not follow this model. The reason for this is because of the settling of sediment in front of the first 0.7 m of Fescue-FS and B-Fescue-FS. This sediment settling was too high to fit data collected below this point. Exclusion of the data from the ponding area and top edge of the grass improved the exponential regressions significantly (p < 0.01). However, regressions are far below the incoming sediment load at –1 m, confirming the dominant role of the upslope edge of the grass with associated ponding on sediment retention of grass barriers and even fescue filter strips. Sediment in runoff decreased exponentially with increasing width of Fescue-FS (r2 = 0.99; Fig. 2b), but decreased linearly in B-Fescue-FS and B-Native-FS (r2 = 0.98). The linear response in B-Fescue-FS is likely due to the reduction of sediment in the ponding area above the barriers.
Nitrogen and Phosphorus Reduction
Nutrient load passing the sampling positions in the Fescue-FS or B-Fescue-FS are shown in Tables 2 and 3. Reduction of organic N, NO3–N, NH4–N, particulate P, and PO4–P was significantly different between the Fescue-FS vs. B-Fescue-FS treatments at 0.7 m (p < 0.01; Table 2). Fescue-FS was less effective than B-Fescue-FS in reducing nutrients in runoff. Values in Fig. 3a
indicate that the 0.7-m Fescue-FS reduced 55% of organic N, 36% of particulate P, 27% of NO3–N, 19% of NH4–N, and 37% of PO4–P when compared with the CCF treatment. In contrast, B-Fescue-FS for equal width reduced 67% of organic N, 53% of particulate P, 68% of NO3–N, 50% of NH4–N, and 54% of PO4–P. Mass of nutrients leaving the 0.7-m Fescue-FS was significantly higher than that leaving the B-Fescue-FS (Fig. 3a). The increased effectiveness of B-Fescue-FS supports a study reporting that 0.8-m wide switchgrass barriers reduced 57% of organic N, 81% of NH4–N, 33% of NO3–N, and 68% of particulate P losses (Eghball et al., 2000). Fescue-FS reduced significantly more NH4–N, NO3–N, and PO4–P than B-Fescue-FS at 4 m (p < 0.01; Table 2). The 4-m strip reduced approximately 58% of nutrients in Fescue-FS and approximately 71% of nutrients in B-Fescue-FS. These results show that the greatest reduction in Fescue-FS occurred between the 0.7 and 4 m, whereas the greatest nutrient reduction in B-Fescue-FS occurred above 0.7 m.
|
|
|
|
Reduction of organic N (r = 0.98) and particulate P (r = 0.99) was positively correlated with trapped sediment. Increased removal of soluble nutrients through B-Fescue-FS at 0.7 m may be due to higher infiltration and adsorption to organic matter, and clay. Infiltration in Mexico silt loam claypan soils is limited under wet conditions because of the slowly permeable Bt horizon; however, infiltration of soluble nutrients near barrier roots penetrating into the Bt horizon likely increase. As discussed, runoff from Fescue-FS was greater than from B-Fescue-FS (p < 0.05) at 0.7 m, suggesting that more runoff infiltrates into the barriers. In fact, Schmitt et al. (1999) on a study on a 6% sloping Sharpsburg silty clay loam reported that switchgrass barriers reduced loss of N and P by increasing infiltration within the barriers. Runoff infiltration in the ponded area above barriers is likely reduced by the very slowly permeable Bt horizon (Blanco-Canqui et al., 2002), thus any increased infiltration would be confined to the barrier zone whose deep roots most likely penetrated the Bt horizon. The reduction of NH4–N, NO3–N, and PO4–P in the fescue strips of the treatments may be caused by (i) adsorption by clay particles and plants, and (ii) infiltration of runoff with colloidal particles (Chaubey et al., 1995). Additional pathways for NO3–N, NH4–N, and PO4–P reduction may include immobilization and biological and chemical transformation (Groffman et al., 1991).
As with sediment, nutrient transport decreased abruptly in the 0.7 m particularly in B-Fescue-FS and B-Native-FS producing a poor exponential regression. Exclusion of the data above the 0.7 m improved the regressions (Fig. 3b through Fig. 5b). Nutrients decreased exponentially with width in Fescue-FS (r2 > 0.96) in accord with Foster (1982). Evaluation of graphs showed that organic N, particulate P, and NO3–N decreased linearly with distance in B-Fescue-FS (r2 > 0.96; Fig. 3b, 4). This is attributed to the large reduction of N and P in sediments above the B-Fescue-FS. The NH4–N and PO4–P decreased gradually with distance of B-Fescue-FS and B-Native-FS following an exponential response (r2 = 0.99; Fig. 5). The exponential decrease of soluble forms of N and P agrees with Srivastava et al. (1996) who showed that N and P were reduced exponentially with Fescue-FS width on a Captina silt loam. In contrast with the sharp decrease of nutrients in the 0.7 m of B-Fescue-FS, Fescue-FS reduced runoff nutrients gradually below the source area.
Prediction of Sediment and Nutrient Removal
Measured and predicted sediment and nutrient trapping of Fescue-FS and B-Fescue-FS are compared in Fig. 6
. The equations by Haan et al. (1994) were used to predict sediment and nutrients. Predicted values for Fescue-FS alone agreed moderately well with measured data (Fig. 6a). Some deviation occurred at higher levels of trapping where the equations slightly underestimated measured values. Linear regression explained 76% of the variance. Agreement between predicted and measured values for the B-Fescue-FS was less adequate than that for Fescue-FS alone (r2 = 0.44). The equations greatly underestimated sediment trapping. Results suggest that applicability of the equations by Haan et al. (1994) is not recommended for use with B-Fescue-FS. Poor performance of the equations is attributed to runoff ponding above the barrier that fails to account for deposition above the barriers. Dabney et al. (1995) and Deletic (2001) stated that current equations only hold for conditions where no runoff-ponding above grass strips occurs. The equations are developed based on the trapping mechanisms of filter-strips rather than barriers. An equation developed by Foster (1982) was used to account for sediment deposition in the ponded area upslope of the barriers:
 | [9] |
is the deposition coefficient estimated from experimental data ( = –0.6) based on the pond length (Lp) above the barrier and sediment entering and leaving the pond. Sediment leaving (qso) the ponded area was then used in Eq. [3] instead of qsi-total. This adjustment improved the regressions between measured and predicted values (r2 = 0.66; Fig. 6b). These results indicate that adjustment for the runoff ponding is critical for prediction of sediment deposition in barriers. However, predicted values were generally lower than observed values at high values.
|
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Switchgrass barriers in combination with vegetative filter strips show promise as a conservation tool for reducing sediment and nutrient loss in runoff and complement current conservation practices.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication September 5, 2003.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. W. Simpson, A. N. Sharpley, R. W. Howarth, H. W. Paerl, and K. R. Mankin The New Gold Rush: Fueling Ethanol Production while Protecting Water Quality J. Environ. Qual., March 1, 2008; 37(2): 318 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Yihe Lu, Bojie Fu, Liding Chen, Guohua Liu, and Wei Wei Nutrient transport associated with water erosion: progress and prospect Progress in Physical Geography, December 1, 2007; 31(6): 607 - 620. [Abstract] [PDF]
|
|
|
|
 |
|
 J. Hussein, H. Ghadiri, B. Yu, and C. Rose Sediment Retention by a Stiff Grass Hedge under Subcritical Flow Conditions Soil Sci. Soc. Am. J., August 9, 2007; 71(5): 1516 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Blanco-Canqui, C. J. Gantzer, and S. H. Anderson Performance of Grass Barriers and Filter Strips under Interrill and Concentrated Flow J. Environ. Qual., October 27, 2006; 35(6): 1969 - 1974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Blanco-Canqui, C. J. Gantzer, S. H. Anderson, and A. L. Thompson Soil Berms as an Alternative to Steel Plate Borders for Runoff Plots Soil Sci. Soc. Am. J., September 1, 2004; 68(5): 1689 - 1694. [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 | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
