|
Effects of Cover Crop and Filter Strips on Sediment and Nutrient Loads Measured at the Edge of a Commercial Cotton Field
Arjun Thapa1, Niroj Aryal1,*, Michele L. Reba2, Tina Gray Teague3, Geoffrey K. Payne2, Anna Pieri4
Published in Journal of the ASABE 67(2): 475-491 (doi: 10.13031/ja.15676). 2024 American Society of Agricultural and Biological Engineers.
1 North Carolina Agricultural and Technical State University, Greensboro, North Carolina, USA.
2 Delta Water Management Research Unit, USDA ARS, Jonesboro, Arkansas, USA.
3 Division of Agriculture, Arkansas State University, Jonesboro, Arkansas, USA.
4 Division of Agriculture, University of Arkansas, Jonesboro, Arkansas, USA.
* Correspondence: naryal@ncat.edu
Submitted for review on 17 May 2023 as manuscript number NRES 15676; approved for publication as a Research Article and as part of the Soil Erosion Research Symposium Collection by Associate Editor Dr. Matthew Helmers and Community Editor Dr. Kyle Mankin of the Natural Resources & Environmental Systems Community of ASABE on 21 November 2023.
Mention of company or trade names is for description only and does not imply endorsement by the USDA. The USDA is an equal opportunity provider and employer.
Highlights
Abstract. Effective use of conservation practices in agricultural fields can reduce sediment and other pollutant loads entering waterways. In this study, we evaluated the effectiveness of using cover crops and filter strips on sediment and nutrient loss at the edge of paired, 7.83 ha (19.35 ac), commercial cotton fields in the Lower Mississippi River Basin (LMRB) in northeastern Arkansas. Cover crops included winter wheat, black oat, and ryegrass seeded in the winter fallow period, while filter strips included a grassy turn row at the field border and switchgrass transplanted around the drainage pipe at the edge of the treatment field. The field border of the control field was generally free from vegetation. A monitoring system measured discharge and collected composite water samples from rainfall and irrigation runoff events. Water samples were analyzed for phosphate (PO4-P), total phosphorus (TP), nitrate (NO3-N), ammonium (NH4-N), total nitrogen (TN), and suspended sediment concentrations. Data were collected from 2015 to 2020. Baseline data were collected when both fields had similar conservation practices (i.e., no cover crops in 2015 and no filter strips in 2015 and 2016). A comparison of 66 common runoff events between the control and cover crop treatment fields during the non-growing season indicated that the median peak flow and sediment loads were significantly reduced (p < 0.05), with an average reduction of 49% and 43%, respectively. Similarly, a comparison of 55 common runoff events between the control and filter strips treatment fields during the growing season found that the filter strips reduced significantly with average runoff by 36%, peak flow by 49%, and sediment loads by 56% (p < 0.05). Nutrient load reductions by the cover crop and filter strip treatments were not significantly different than by the control (p > 0.05). However, mean PO4-P, TP, NO3-N, NO2-N, NH4-N, and TN loads in the cover crop treatment field were lower than in the control field by 21%, 4%, 9%, 4%, 17%, and 7%, respectively. Similarly, mean PO4-P, TP, NO3-N, NH4-N, and TN loads in the filter strip treatment field were lower than in the control field by 23%, 15%, 11%, 42%, and 21%, respectively. The results demonstrated runoff depth, peak flow rate, nutrients, and sediment load reductions following the implementation of cover crops and filter strips at the commercial field scale.
Soil tillage, use of inorganic fertilizers, and high irrigation-water usage, each intended to increase crop production, lead to soil erosion and nutrient losses from agricultural fields. Sediment and nutrients from fields are deposited in river systems, lakes, and coastal areas worldwide, contributing to poor ecosystem conditions. In the United States, approximately 46% of rivers and streams, 32% of wetlands, and 18% of coastal water are considered to be in poor biological condition (USEPA, 2017). Agriculture’s contribution to non-point source pollution is well documented. For example, Robertson and Saad (2021) estimated that agricultural sources contributed 60% nitrogen and 56% phosphorus to the Mississippi and Atchafalaya River Basins. The compounding issues are continued eutrophication in the Mississippi River Basin, resulting in harmful algal blooms, and a large hypoxic zone with fish kills in the Gulf of Mexico annually (Thrash et al., 2017). In 2020, the midsummer areal extent of the hypoxic zone in the Gulf of Mexico was 5,048 square kilometers, and nutrient loads were the highest in May (USEPA, 2022). Eutrophication is estimated to lead to a $2.2 billion annual loss to the US economy (Dodds et al., 2009).
The Mississippi River Delta region of Arkansas is an economically important region for US agricultural production. In 2021, agriculture occupied 42% of the state's total land area (5.65 million hectares out of 13.47 million hectares, or 33.3 million acres). Rice, cotton, soybean, and corn are the principal row crops cultivated in the region (Kassel, 2023). In 2022, the agriculture sector contributed approximately $19 billion to the Arkansas economy. However, farm management practices such as tillage, irrigation, and fertilizer applications can result in excessive irrigation and precipitation runoff, erosion, and nutrient losses (Kaushal et al., 2011). The adoption and implementation of conservation practices are needed to reduce the footprint of the region’s agriculture on the environment. To accomplish this, research-based, practical conservation practices for row crop production have been developed and applied to mitigate nutrient and sediment losses from agricultural fields and improve water quality. Examples of these cost-effective and ecologically sound technologies include crop rotation, cover cropping, streamside vegetative buffers, filter strips, modified nutrient and irrigation management, conservation tillage, and denitrifying bioreactors (Centner et al., 1999; Christianson et al., 2021).
The use of cover crops (USDA NRCS Conservation Practice Code 340) and filter strips (USDA NRCS Conservation Practice Code 393) was found to be effective in reducing sediment, nitrate, total nitrogen, dissolved phosphorus, and total phosphorus from surface runoff (Aryal et al., 2018; Kaspar and Singer, 2011; Meisinger et al., 1991; Yuan et al., 2022). Cover crops and filter strips reduce the sediment and nutrient load in the runoff through infiltration, adsorption, plant uptake, sedimentation, metabolization, and degradation (Sharply et al., 1991; Lee et al., 1998; Rahman et al., 2013). Cold-resistant cover crops such as ryegrass and winter wheat reduce runoff, sediment, and nutrient load in freshwater bodies during spring growth (Sharply et al., 1991; Odhiambo et al., 2012). They also mitigate wind erosion and conserve soil moisture (Woodruff, 1972). Similarly, switchgrass grows well in the warm and humid climates of Arkansas and has been effective at reducing sediment and nutrient loads in agricultural fields at the plot-scale (Lee et al., 2000; Sanderson et al., 2012). Switchgrass is an excellent candidate for biofuel production due to its biomass and suitability for growing on marginal land (Kim et al., 2023). Additionally, tall fescue, known for its winter tolerance capacity, is successful as filter strips in turn rows. Tall fescue excels in reducing nutrient and sediment loads during the transition from cover crop termination to switchgrass establishment (Lee et al., 1998; Blanco-Canqui et al., 2006; Sheng et al., 2021).
Research validation studies at the commercial field scale are needed to understand the effects of single or multiple conservation practices on hydrology and the dynamics of nutrients and sediment. While plot scale studies on conservation practices such as cover crops and filter strips have been conducted (White and Arnold, 2009), these results may not fully represent sediment and nutrient loads from commercial agricultural land (Dabney, 1998; Reba et al., 2020). This study addresses the knowledge gap by assessing how cover crops and filter strips perform in reducing nutrient and sediment loads in commercial cotton fields in the Lower Mississippi River Basin (LMRB). The objective was to evaluate the effectiveness of cover crops and filter strips in mitigating hydrology, sediment, and nutrient loads at the commercial cotton field.
Materials and Methods
Study Sites
The study was conducted from 2015 to 2020 at a commercial cotton farm in Mississippi County, near Manila, AR, in the LMRB (fig. 1). The site was located in the Little River Ditches Watershed (8-digit HUC-08020204), a USDA NRCS prioritized watershed. At the start of the study, one large field was split into two parts and identified as a treatment and control field, each 7.83 ha (19.35 ac). Soils in the fields were approximately 67.9% Routon-Dundee-Crevasse complex and 31.1% Sharkey-Steele complex soil with less than 1% slopes with hydrologic groups C/D and D, respectively. Two-thirds of the field had silt loam and silty clay loam, and the remaining one-third had silty clay and clay. According to the Web Soil Survey accessed on 1 Feb 2023, the proportion of each soil type was similar across both treatment and control fields (fig. 1).
Precipitation was measured using a Campbell Scientific weather station located approximately 400 meters west of the study fields, near the producer’s house. This location allowed the producer to easily inspect equipment and report any damage or malfunction of equipment. Precipitation data from the tipping bucket rain gauge was post-processed to calculate daily, event-wise, monthly, and yearly totals based on time-stamped rainfall data. The cover crops spanned from fall to winter, whereas the cotton growing season extended from May to October. Rain from June to August played a vital role in scheduling irrigation and cotton crop yield. Year-round rainfall measurement allowed for establishing a relationship between rainfall and sediment and calculating the nutrient load under cover crop and filter strips.
Crop Management
Cotton was planted in the fields every year of the study period from 2015 to 2020. Cotton was typically planted in the first or second week of May and harvested in October (growing season). The specific cotton growing season for each year during the study period is provided in table 1. Before cotton planting, the tops of the beds were lightly flattened with a field cultivator equipped with a rolling basket. Cotton was planted in early to mid-May with a seeding rate of approximately 9 seeds per m-1 on raised beds with 1 m row spacing. The cover crop was seeded using a Gandy Ortho Air seeder mounted on the disk bedder after reshaping the beds after cotton harvesting. It was terminated one or two weeks before cotton planting and the period between seeding and termination was considered the cover crop period (Treatment CC). Pre- and post-emergent herbicides were applied during the production season to control the cover crop and weeds present in the field. Phosphorus (P) and potassium (K)-based fertilizers were typically applied before planting, while nitrogen (N)-based fertilizer was applied after planting (table 1). The N-based fertilizer was broadcast as a top dress in either a single application or as multiple split applications over time. There were also foliar applications of N fertilizer during the season. Furrows were cleared four to six weeks after planting before the first irrigation using an Orthman cultivator (middle busters). Typically, the first irrigation was applied around mid-June, during the first or second week of the square (first appearance of floral buds). Furrow irrigation was delivered using polyethylene irrigation tubing (poly pipe) with groundwater from the Mississippi River Valley Alluvial Aquifer. Irrigation was applied multiple times throughout the growing season based on weather conditions and crop demand. Irrigation is typically terminated in late August based on soil moisture availability and plant maturity (Teague and Reba, 2014). Most of the irrigation event data were recorded and reported by the farmer or were based on flow meter readings. At the end of the cropping cycle, plant growth regulators and defoliants were applied two weeks before harvesting. Cotton was typically harvested before November (table 1). After harvesting, stalks were shredded and left on the soil surface, accounting for 10%-20% cotton residue.
 |
| Figure 1. Location of the study site. Reference map shows the state, watershed, county, and field scale (sky blue line is the field divider, and green circles are the field drainage locations on the east side of the fields in Mississippi County, AR; 35°51’51.9”N, 90°14’28.5”W). |
| Table 1. Summary of crop management and agronomic practices for the study period. | |||||||||
| Year | Cover Crop | Filter Strips | Tillage | Crop | Planting Date | Fertilization Date | Fertilizer (kg ha 1); Application Method | Furrow Runner/ Furrow Plow | Harvesting Date |
| 2015 | None | None | 2-May | Cotton | 2-May | 2-Jun | N (112); broadcast | 24-Jun | 22-Oct |
| 9-Jun | P (22.4), K (67.2); broadcast | ||||||||
| 2016 | Winter wheat | None | 8-May | Cotton | 8-May | 4-Apr | P (61), K (99); variable rate | 13-Jun | 28-Oct |
| 6-Jul | N (2.95); split and foliar | ||||||||
| 20-Jul | N (11.77); spilt and foliar | ||||||||
| 2-Aug | N (11.77); spilt and foliar | ||||||||
| 2017 | Winter wheat | Fescue + Switchgrass | 10-May | Cotton | 10-May | 10-May | P (61), K (99); variable rate | 18-Jun | 26-Oct |
| 26-Jul | N (11.2); split and foliar | ||||||||
| 2-Aug | N (0.06); split and foliar | ||||||||
| 2018 | Ryegrass | Fescue + Switchgrass | 2-May | Cotton | 2-May | 11-Jul | N (112); broadcast | 15-Jun | 19-Sep |
| 2019 | Ryegrass | Fescue + Switchgrass | 15-May | Cotton | 15-May | 15-May | P (22.4), K (90); broadcast | 24-Jun | 23-Oct |
| 11-Jun | N (90); broadcast | 23-Oct | |||||||
| 2020 | Black oat | Fescue + Switchgrass | 2-May | Cotton | 11-May | 11-May | N (112); broadcast | 5-Jun | 8-Oct |
In 2015, a baseline scenario was evaluated by following the same conventional practices in both fields to evaluate their inherent differences. Beginning in 2016, an annual spring cover crop of either wheat (Triticum aestivum), cereal ryegrass (Secale cereale), or black oats (Avena strigose) was planted in the treatment field (table 1). Along with the cover crop, a filter strip was constructed in the spring of 2017 by transplanting mature switchgrass (Panicum virgatum) plugs in a 2 m radius semicircle (total area ~12 m2) around the discharge pipe in the treatment field. The turn row area of the treatment field bottom was planted with Kentucky 31 tall fescue (Festuca arundinacea Schreb) via hand broadcast to complete the filter strip (fig. 2b). The runoff through filter strips treatment between cotton crop plantation dates and harvest dates was called Treatment FS. Establishment of tall fescue in the filter strip area was challenging due to herbicide damage and variable weather conditions. The filter strip was maintained through 2020. This maintenance included additional switchgrass transplants in 2018. In the control field, emergent vegetation at the field or front of the drainage pipe was mowed or treated with herbicides, as is the conventional practice in the study area.
 |
| Figure 2. Illustration of cotton edge-of-field drainage areas (inflow) with varying degrees of vegetation coverage in front of drainage pipe and turn row, including (a) the control field, (b) treatment field, and (c) ideal coverage of multiple vegetation types when establishment of filter strip was successful. |
Water Sampling Instrumentation
The study was designed according to the USDA-NRCS National Activity Standard for Edge-of-Field (EOF) Monitoring, which includes Water Quality Monitoring Data Collection and Evaluation (CEMA 201) and EOF Water Quality Monitoring System Installation (CEMA 202). An automated ISCO 6712 sampler (Teledyne ISCO, Lincoln, NE) was set up near the discharge pipe of each field (green dot in fig. 1). An ISCO 2150 area-velocity module and area-velocity sensor (Teledyne ISCO, Lincoln, NE) were mounted to a steel instrument sled and placed inside a 0.609 m (2 ft) diameter surface drainage pipe to measure discharge and flow. All sensors were connected to a data logger (CR1000, Campbell Scientific, Logan, UT) for data recording. Two radios (RF401, Campbell Scientific, Logan, UT) and a cell modem (Raven RXT, Campbell Scientific, Logan, UT) transmitted data from the field to be stored on a server at Arkansas State University using the AT&T GSM Network. All equipment and sensors were powered by a 12V deep cycle battery (Interstate SRM-27) and a 120-watt solar panel with a 12-volt, 4.5-amp regulator. The water sampler, charger, regulator, battery, data logger, cell phone modem, and transmitter were enclosed in a weather-resistant enclosure, as shown in figure A1 in the appendix. More information regarding instrumentation details can be found in Aryal et al. (2018).
The composite water sample from each rainfall or irrigation runoff event was sampled at a rate of 400 mL every 50 m3 of runoff and stored in a 10-L sample bottle. The sampler was programmed to collect uniformly throughout an event on a flow basis, including both the rising and falling limbs of the hydrograph. When a runoff event began, sensors recorded water depth (m) and velocity (m/s), including a maximum and average recording stored on the datalogger every 15 minutes. When the water depth reached 0.02 m, the datalogger began to count the discharge to the 50 m3 trigger volume. Once a sample trigger occurred, an alert was sent via text to a technician. The sample was collected within 24 hours of the alert and transferred, on ice, to a laboratory for sediment and nutrient analysis. Samples collected between 2015 and 2018 were delivered to the Ecotoxicology Research Facility (ERF) at Arkansas State University, while those collected between 2019 and 2020 were delivered to the Water Quality Research Lab (WQRL) at the USDA-ARS-Delta Water Management Research Unit. The field site was regularly maintained, with visits every two weeks or during sample collection. During maintenance, a technician inspected equipment, calibrated the sample collection volume, and restocked necessary supplies as needed. The technician also visited as soon as possible when a low battery message was received from the field.
Water Quality Analysis
Suspended sediment concentration (SSC) was measured using ASTM Method D3977-97, total suspended solids using APHA 2540-D; nitrite N (NO2-N) and nitrate N (NO3-N) using USEPA 353.2, APHA 4500-NO3I, and APHA 4500-NO2B; total N (TN) using USEPA 350.1 and APHA 4500-P J, phosphate P (PO4-P) using USEPA 354.1 and APHA 4500-P F; and total P (TP) using USEPA 365.4 and APHA 4500-P J (American Public Health Association, 2005). The detection limit was 0.01 mg L-1 for PO4 and TP, 0.04 mg L-1 for NO3 and TN, and 0.002 mg L-1 for NO2 at ERF. More information regarding water quality analysis can be found in Reba et al. (2020). TN and TP were analyzed by alkaline persulfate digestion with molybdenum blue colorimetric analysis method for TP and zinc reduction method for TN (American Public Health Association, 2005). Ammonium nitrogen (NH4-N) was measured by the indophenol blue (IPB) method (American Public Health Association, 2005). The nitrate and nitrite were measured by the vanadium (III) reduction method (Doane and Horwáth, 2003). At WQRL, the detection limit was 0.01 mg L-1, following Adviento-Borbe et al. (2018). The sediments TP, PO4-P, TN, NO3-N, and NO2-N were started to be analyzed from 2015 to 2020, but NH4-N was analyzed from 2019 to 2020.
For quality assurance and quality control, calibration curves were made with each reagent batch with an R2 value of at least 0.997. Additionally, blanks, duplicate runs, and check standards were used every 10 samples for detecting possible contamination and instrument drift. In addition to the above Quality Control (QC) measures for nutrients, a digestion efficiency QC sample (nicotine acid for N and adenosine triphosphate for P) was included at a frequency of 1 per 20 samples. These QC samples underwent the entire digestion procedure and were maintained within ±10% of the known concentration. Furthermore, a laboratory control sample was included from the beginning to the end of each run, and its concentration was kept within ±10% of the known concentration.
Data Analysis
The total discharge per runoff event was calculated as the sum of 15-minute calculated discharges from water depth and velocity data collected from each field. The concentrations of sediment and nutrients in runoff water were obtained from laboratory analysis. The total load was calculated by multiplying the total discharge by measured concentrations of analytes in runoff water. The load per unit area was calculated by dividing the total load by the drainage area. The percentage load reduction was calculated by the load in control minus the load in treatment divided by the load in the control and multiplied by 100. All data calculations and formatting were completed in Excel (Microsoft, Redmond, WA). Non-normal data were examined at a = 0.05 using the Mann-Whitney Rank Sum Test for median comparison in Sigma Plot version 15.0.
 |
| Figure 3. Annual rainfall at the site in Mississippi County, AR, for all study years compared with 30-year normal rainfall. |
Results and Discussion
Precipitation
Annual rainfall varied from 1204 mm to 2506 mm, with the lowest and highest annual rainfall recorded in 2016 and 2019, respectively (fig. 3). Rainfall data indicated that 2016 and 2017 were drier years, while 2018 and 2020 were wetter years compared to the 30-year normal rainfall. Monthly rainfall data revealed that the six-year average monthly rainfall was lowest in September (22.4 mm) and highest in February (189.8 mm). The monthly average data also demonstrated that the cotton growing season (summer) experienced less rainfall compared to the cover crop period (winter; fig. 4). During the study period, the highest and lowest monthly rainfall occurred in February 2019 and September 2017, respectively (fig. 5).
Baseline Assessment
During the baseline period for cover crops in 2015, the cover crop was seeded neither in the treatment (Baseline T CC) nor in the control field (Baseline C CC), and measurements were taken for eight common runoff events. There were no significant differences in runoff depth (mm), peak flow rate (m3s-1), sediment load (kg ha-1), and nutrient load (kg ha-1) between the treatment and control fields during the baseline period (figs. 5-6, 9, 11-12, and A2-A4). The baseline period for filter strips was the growing seasons of 2015 and 2016, and neither switchgrass nor fescue were planted in the treatment field (Baseline T FS) nor in the control field (Baseline C FS). Rainfall or irrigation generated four common runoff events measured in both the treatment and control fields in 2015 and three common runoff events in 2016. There were no significant differences in runoff (mm), peak flow rate (m3s-1), sediment load (kg ha-1), and nutrient load (kg ha-1) between the treatment and control fields during that period (figs. 5-6, 9, 11-12, and A2-A4). These data implied that there were no inherent differences between the two fields.
Effectiveness of Cover Crops and Filter Strips
Cover crops were typically planted in the fall after cotton crop harvest and terminated in the spring before the cotton crop was planted, referred to as Treatment CC, and without cover crop treatment, called Control CC. The specific schedule for each year's cover crop is detailed in table 1. Water quantity and quality data from 66 common runoff events from 2016-2020 were used for cover crop evaluation. In 2016, nutrient and sediment load analysis during the cover crop season was based on only three runoff samples due to 10 missed discharge events (the lowest recorded common runoff data among treatment years). The time period between when the filter strip was grown during the cotton season is referred to as Treatment FS, and without filter strip, it is referred to as Control FS. Water quality and quantity data from 55 common runoff events (comprising 27 irrigation and 28 rainfall events) were used for the filter strip evaluation from 2017 to 2020.
 |
| Figure 4. Monthly rainfall at the site in Mississippi County, AR, for the study years compared with 30-year normal rainfall. |
 |
| Figure 5. Runoff per event during the growing seasons of 2015 to 2020. The red lines indicate the mean, and horizontal black lines specify the 10th, 25th, median, 75th, and 90th percentile. Different letters within the same year indicate significant differences. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
Hydrology
In 2018, the median runoff depth was significantly reduced (p = 0.043) by the cover crop (Treatment CC); however, the mean runoff depth decreased in all treatment years (fig. 5). When considering the combined common runoff events for the cover crop treatment, it was evident that both the mean and median runoff depth for the treatment field were lower than those of the control, although the median runoff depth reduction was not statistically significant compared to the control (p = 0.05). The statistically insignificant reduction in runoff depth from the cover crop treatment field may be due to either partial or late cover crop germination due to adverse weather conditions affecting seed germination or planting in most years. The per-event average runoff depth from the cover crop treatment and control field was 7.04 mm and 10.09 mm, respectively, resulting in a 30% reduction in runoff depth by the cover crop.
 |
| Figure 6. Maximum flow per event during the growing seasons of 2015 to 2020. The red lines indicate the mean, and horizontal black lines specify the 10th, 25th, median, 75th, and 90th percentile. Different letters within the same year indicate significant differences. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
In 2017 (p = 0.046) and 2018 (p = 0.002), the filter strips significantly reduced the median runoff depth compared to the control (fig. 5). However, the median runoff depth was not significantly reduced by the treatment compared to the control in 2019 and 2020, even though the mean runoff depth was lower in the treatment. The lack of significant runoff reduction in the treatment field in 2019 and 2020 may be due to the greater amount of event-wise rainfall intensity and duration (graphs are not shown here) that was reflected in higher monthly and yearly average rainfall (2506 mm in 2019 and 1800 mm in 2020) during the growing season compared to the 30-year normal rainfall (1277 mm; fig. 3). Combining common runoff (27 irrigation and 28 rainfall) events from the growing seasons of 2017 to 2020, the median runoff depth was significantly lower in the treatment than in the control (p = 0.003). The treatment resulted in a 36% reduction in average runoff depth (6.11 mm) per event compared to the control (9.56 mm).
The reduction of runoff depth could be attributed to factors including increased infiltration and transpiration from cover crops (Blanco-Canqui et al., 2011; Dabney et al., 2001; Reeves, 2018; Gavric et al., 2019) and increased infiltration and evaporation from the short-term ponding induced by switchgrass filter strips near the drainage pipe inlet. The presence of grassroots in the treatment field also makes the soil more porous, thereby increasing infiltration compared to the bare soil surface, which enhances evapotranspiration and surface retention. (Deletic, 2000). However, some cracking in the bare soil surface (Cheng et al., 2021) and sparse grasses presented in the control field may have also contributed to the increased infiltration and evapotranspiration, potentially underestimating the true effectiveness of cover crops and filter strips.
The cover crops significantly reduced the median peak flow rate in 2018 (p < 0.001), 2019 (p = 0.015), and 2020 (p = 0.007) compared to the control. Combining the common peak flow rate per event from 2016 to 2020 (N = 66), the cover crop treatment exhibited a significantly lower peak flow rate than the control (p <0.001; fig. 6). The combined common per-event average peak flow rate for the treatment and control was 0.023 m3s-1 and 0.045 m3s -1, respectively, resulting in a 49% reduction in peak flow rate. Across common runoff (27 irrigation and 28 rainfall) events during growing seasons from 2017 to 2020, the median peak flow rate was significantly lower in the treatment than in the control (p < 0.001; fig. 6). The average peak flow rate per event in the treatment and control was 0.022 m3s-1 and 0.044 m3s-1, respectively, a 49% reduction in the filter strip treatment versus the control. The significant reduction in median peak flow rate in the treatment was due to decreased runoff velocity, increased time of concentration by the cover crops, and hydraulic retention by the filter strips as a result of higher hydraulic resistance caused by vegetation (Deletic, 2000; Winston et al., 2019). During the study period, the highest peak flow due to rainfall occurred in the control field on August 31, 2017 (0.11 m3s-1) in filter strips due to 96 mm of rainfall and on 31 October 2018 (0.12 m3s-1) in the cover crops field due to 27 mm rainfall. Even though the rainfall on 31 October 2018 was lower than that on 31 August 2017, the peak flow rate was higher as the cotton crop was already harvested and the cover crop had not grown well to resist the flow.
The cover crop and filter strip treatment had a significant effect on flow depth and peak flow rate on an event basis. A typical hydrograph (fig. 7a and 7b) demonstrated how cover crops and filter strips reduced the peak flow and delayed the runoff. The flow rate in the treatment field lagged and flattened compared to the control field. Reduced peak flow and runoff volume, longer time of concentration, and increased retention time are critical for erosion protection, nutrient reduction, and reduced downstream flooding (Marttila and Kløve, 2009; Hernandez-Santana et al., 2013). These factors also mitigate sediment-laden pollutant transport to water resources.
  | |
| (a) | (b) |
| Figure 7. (a) Example of a typical hydrograph from the treatment and control field during the cover crops on 17 Feb, 2017. (b) Example of a typical hydrograph from the treatment and control field during the growing season on 26 Aug, 2019. | |
Sediment
In 2019, the annual median sediment load was significantly reduced by cover crop treatment compared to control (p = 0.015; fig. 8). Although the median sediment load reduction by the cover crop treatment was not statistically significant in years other than 2019 when compared to the control, the mean sediment load reduction was consistently higher by the cover crop treatment throughout all treatment years. When considering the combined common per-event runoff from 2016 to 2020, it was evident that the median sediment load was significantly reduced by cover crop treatment compared to control (p = 0.008; N = 54). The average sediment load per event from cover crop treatment and control was 56.12 kg ha-1 and 98.98 kg ha-1, respectively, resulting in an average 43% reduction in sediment load by the cover crop.
The annual median sediment load was significantly reduced by the filter strip treatment during the growing season of 2018 (p = 0.003). The mean sediment load reduction by filter strip was higher than that by control in each year during the treatment period. The median sediment load of the combined common runoff (25 irrigation and 26 rainfall) events in growing seasons from 2017 to 2020 was significantly lower in the filter strip treatment than in the control (p = 0.027; fig. 8). Average sediment load per event was 56% lower in the treatment (26.44 kg ha-1) compared to the control (59.76 kg ha-1). The maximum sediment load from the treatment and control fields were 103.6 kg ha-1 and 1023.48 kg ha-1, respectively, on May 11, 2017. Results obtained from 2015 to 2020 showed sediment load had a positive linear relationship with the runoff (r2 = 0.32; N = 307). The presence of densely populated switchgrass filter strips and cover crops reduced runoff depth, flow velocity, and peak flow rate by increasing infiltration, evapotranspiration, and resisting free flow through vegetation, ultimately trapping sediment from the runoff generated in the cropping field (Dillaha et al., 1988; Deletic, 2000; Korucu et al., 2018; Winston et al., 2019).
The sediment load decreased as the season progressed and plant growth occurred (fig. 9). As plants grow and root length increases, roots bind soil particles tightly with their root network and increase infiltration by micro- and macropore formation, which helps to reduce soil erosion and sediment load from the field (Vannoppen et al. 2015). Mature plants also provide a canopy to the soil surface that reduces the kinetic energy of rain droplets impacting the soil surface. The higher sediment load at the beginning of the growing season was attributed to soil surface disturbance by tillage operations and poor plant cover due to cotton or switchgrass filter strips being in the early stages of development (Dendy, 1981). In the growing season, most of the early runoff was generated by rainfall, and the late runoff was generated by irrigation. Figure 9 shows that the sediment load in the early growing season was higher than that of the late season, which also indirectly implies that the sediment load from rainfall was higher than that from irrigation. Sediment load depends heavily on slope, vegetative cover, and soil type (Defersha et al., 2011). Lambrechts et al. (2014) reported that sediment retention efficiency by ryegrass (L. perenne) filter strips increased from 35% to 50% from two to four months of growth due to plant morphology development and tillering capacity of plants.
 |
| Figure 8. Sediment load per event during the growing seasons of 2015 to 2020. The red lines indicate the mean, and horizontal black lines specify the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
 |
| Figure 9. Typical sediment load trends from treatment and control fields during the growing season in 2020 as the season progressed. |
Nutrients
The median TP load was not significantly reduced by cover crops in all treatment years. However, the mean TP load per event from the treatment field was lower than from the control field in all the treatment years except in 2017 and 2020 (fig. 10). The combined common runoff event average TP load in the cover crop treatment and control were 0.08 kg ha-1 and 0.083 kg ha-1, respectively, representing a 4% reduction from 2016 to 2020 (N = 58). Similarly, the combined common runoff event average PO4-P load in cover crop treatment was 0.030 kg ha-1 and in the control field was 0.038 kg ha-1, showing an average reduction of 21% (N = 46; fig. A2). This reduction exhibited a similar pattern to the TP load each year. The median TP load in the filter strip treatment was significantly lower than the control in 2018 (p = 0.025). No significant differences in TP load were identified between the filter strip treatment and control in 2017, 2019, or 2020 (fig. 10). Results obtained from 2015 to 2020 showed that the TP load had a positive linear correlation with the runoff (r2 = 0.27; N = 306) and sediment (r2 = 0.23; N = 306). Since TP is a cumulative measure of dissolved reactive phosphorus and particulate phosphorus, the TP load spiked on May 11, 2017, in the control field due to a higher sediment load caused by a higher peak flow rate and total flow. The PO4-P load also showed similar results as TP and was not significantly different than the control in either treatment (fig. A2).
Combining common runoff (24 irrigation and 27 rain) events during growing seasons from 2017 to 2020, the average TP load in the filter strip treatment was lower than the control. The per-event average TP load for filter strip treatment was 0.038 kg ha-1 and control was 0.044 kg ha-1. Similarly, per event, the average PO4-P load for filter strip treatment and control was 0.021 kg ha-1 and 0.028 kg ha-1, respectively. The filter strip treatment reduced the average TP load by 15% and PO4-P by 23% compared to the control field.
 |
| Figure 10. Total phosphorus load per event during the growing seasons of 2015 to 2020. The red lines indicate the mean, and horizontal black lines specify the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
The median TN load was significantly lower in 2018 (p = 0.028), whereas the average TN load was consistently lower in the cover crop treatment field than in the control, with the exception of 2017 (fig. 11). The combined common runoff event average TN load was lower in the treatment field, but there was no significant difference in median TN load compared to the control (N = 58, P = 0.639). Similarly, both NO3-N and NO2-N loads were significantly lower in the treatment field in 2018 (figs. A3 and A4). Even though the combined common runoff event median NO3-N and NO2-N load was not significantly different, the mean load was lower in the cover crop treatment field than the control. The combined common event average TN load from cover crop treatment and control was 0.080 kg ha-1 and 0.086 kg ha-1, respectively, with a 7% reduction. The combined common runoff event average NO3-N load in the cover crop treatment was 0.020 kg ha-1 and the control field was 0.022 kg ha-1 with a 9% (N = 49) reduction, while the NO2-N load in the treatment was 0.0043 kg ha-1 and the control was 0.0045 kg ha-1 with a 4% (N = 54) reduction. The median NH4-N load from the treatment and control fields was not statistically different; however, the average load was lower in the treatment field for both years (fig. A5). The combined common per-event average NH4-N load in the treatment was 0.0059 kg ha-1 and in the control was 0.0071 kg ha-1, with a 17% reduction.
Although the median TN load was not significantly different for the entire duration of the filter strip treatment period, the mean TN load from the treatment was lower than the control each year (fig. 11). The combined common per-event average TN load from 2017 to 2020 (24 irrigation and 27 rain) for treatment and control was 0.10 kg ha-1 and 0.13 kg ha-1, respectively. The NO3-N load followed similar trends as the TN load during the study period (fig. A3). Both NO3-N and NO2-N loads from the filter strip treatment were not significantly different than the control (figs. A3 and A4). The combined common per-event average NO3-N load was 0.0458 kg ha-1 and 0.0516 kg ha-1 in filter strip treatment and control, respectively. Similarly, the combined common per-event average of NO2-N load was 0.0096 kg ha-1 in the filter strip treatment and 0.0091 kg ha-1 in the control. The median NH4-N load from the treatment and control fields was not statistically different; however, the average load was lower in the treatment field for both years (fig. A5). The combined common per-event average NH4-N load in the treatment was 0.0102 kg ha-1 and in the control was 0.0175 kg ha-1.
Across the combined common runoff events during four growing seasons, the average TN, NO3-N, and NH4-N loads were lower in the treatment than in the control, but the median loads were not significantly different (p > 0.05). Average TN, NO3-N, and NH4-N loads in the treatment were reduced by 21%, 11%, and 42%, respectively, compared to the control field during the treatment period. However, the NO2-N load was 6% higher in the treatment field than in the control field during the same period.
 |
| Figure 11. Total nitrogen load per event each year during the growing season. The red lines indicate the mean, whereas horizontal black lines in the box plot indicate the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
The low percentage reduction of the NO3-N and NO2-N load by cover crop and filter strips was due to the highly soluble nature of these ions in runoff water, which could not be sorbed significantly by vegetation. In addition, these negatively charged ions are poorly adsorbed or immobilized by negatively charged sediment particles. Most of these soluble nutrient loads are reduced through plant uptake. A smaller width of switchgrass partially killed fescue filter strips, and late or partially germinated cover crops with a greater flow rate provided fewer opportunities for plant uptake of the soluble nutrients, leading to less nutrient load reduction in the treatment field (Mitsch et al., 2001). The higher average NH4-N load reduction (17% by cover crops and 42% by filter strips) than TN, NO3-N, and NO2-N percentage load reduction in the treatment field may be due to the attachment of positively charged ammonium ion (NH4+) with negatively charged sediment particles and sedimentation in filter strips and cover crops or NH4+ uptake by plants through their root system or anaerobic microorganisms (Mitsch et al., 2001). Nutrient load reduction by cover crops and filter strips also depends on the antecedent soil moisture content, as it influences the total runoff, peak runoff rate, and compaction of the soil. Cover crop and filter strips reduce significant amounts of nutrient load if the soil is dry by absorbing most of the runoff (Techer and Berthier, 2023). Additionally, the maintenance of cover crops and vegetative filter strips may also affect the sediment and nutrient load, as it will help prevent channeling and short-circuiting water (Rahman et al., 2013).
Over five seasons for cover crops and four growing seasons for filter strips, treatment did not significantly reduce the median P and N load compared to the control. The insignificant reduction in nutrient and sediment load observed in the cover crop treatment may be attributed to the poor and delayed germination of the cover crop caused by extreme cold conditions during the germination period (11 days below 0°C in 2017 and 16 days below 0°C in 2018; data taken from astate.edu). Additionally, the increasingly seasonal nature of rainfall in northeast Arkansas prevented the timely planting of the cover crop for over-winter establishment, resulting in the delayed appearance of the cover crop in the spring. The cover crop treatment reduced combined common per event average loads of runoff depth, peak flow rate, and sediment load by 30%, 49%, and 43%, respectively, and TP, PO4-P, TN, NO3-N, NO2-N, and NH4-N load by 4%, 21%, 7%, 9%, 4%, and 17%, respectively. Aryal et al. (2018) observed a 39% sediment reduction, an 86% NO3-N reduction, and a 53% PO4-P reduction due to cover crops. Additionally, Lee et al. (2016) reported higher reductions in plot studies, with winter rye grass cover crops reducing 49.3% of NO3-N and 91% of K, surpassing the reductions achieved in this study but comparable to the range in runoff volume (10 to 98%) and sediment (22 to 100%) reported by Blanco-Canqui (2018).
After the cover crops were terminated using herbicide application, residue remained on both sides of the furrows, even after being cleared by furrow tillage before irrigation. This residue could have influenced nutrient loads during the crop growing season, although it was not directly measured in this study. Another reason for the insignificant difference in nutrient load reduction between the cover crops or filter strips may be the presence of sparse native grass in the control field turn row, despite herbicide application and mowing. This native grass in the control field in the turn row likely affected nutrient and sediment load reduction. Additionally, Prosser et al. (2020) noted that the width of the switchgrass filter strip used in the treatment was relatively small, constituting only 0.015% of the total cropping area of the field. This minimal proportion could have contributed to the observed insignificant reduction in nutrient load. Though the intent was to provide full coverage of the bottom drainage area (~ 3100 m2) of the treatment field with tall fescue, herbicide drift from field applications and variable weather prevented stand establishment. Another challenge with establishing the wider buffer area was the producer’s perception that the grass would create a drainage problem. Even though filter strip effectiveness depends heavily on soil type, slope, and antecedent soil moisture content, species diversity, size, and density also play an important role (Lambrechts et al., 2014; Prosser et al., 2020). For example, soil stabilization by plants depends on root density, length density, tensile strength, and system morphology in the soil (He et al., 2023). If filter strips have various species diversity and richness, there is a higher chance of binding numerous soil particles from various root zones as well as helping to infiltrate water, sedimentation, and nutrient uptake. The other benefit of having species diversity and richness is that it diversifies the above-ground stem density as well as the survival period of the plant, which provides hydraulic resistance to the flow.
Lee et al. (1998) reported that 3-m switch grass filter strips removed 28% of TN, 42% of NO3-N, 37% of TP, and 43% of PO4-P in 20:1 and 40:1 area-to-length ratios, which was similar to this study. But in the same study, 6-m filter strips removed a significantly greater percentage of nutrients than 3-m filter strips. Similarly, Goel et al. (2004) reported that NO3-N and PO4-P percentage reductions increased significantly when filter strip width increased from 5 m to 10 m. Similarly, after a meta-analysis of 90 filter strip studies used as best management practices, Ramesh et al. (2021) reported that typical vegetative filter strips with a width of 10-20 m show the highest effectiveness in reducing sediment and nutrients. Bortolozo et al. (2015) found that a 10-m wide filter strip was enough to reduce sediment and nutrient loads from runoff, but 30-m wide filter strips performed best for soluble nutrients in a 5 m2 to 30 m2 plot scale study, meaning wider filter strips reduce more nutrient load than narrow filter strips.
Filter strip width also affects the removal of sediment. Dillaha et al. (1988) reported sediment yield reduction by 4.6-m and 9.1-m filter strips was 81% and 91%, respectively, in 5.5 m wide and 18.3 m long plots. Similarly, Lee et al. (1998) reported 66% and 77% sediment reduction by 3-m and 6-m switch grass filter strips, respectively. Filter strips trap sediment particles effectively within 0.5 m width of filter strips until plants are buried in that area and move forward (Dillaha et al., 1988). The 2 m width of filter strips used in the current study was chosen for optimum land utilization. They were effective in sediment load reduction while not sacrificing acreage. Van Dijk et al. (1996) and Abu-Zreig (2001) have reported similar sediment load reductions with 15 x 0.5-m filter strips (Van Dijk et al., 1996; Abu-Zreig, 2001).
Summary and Conclusion
This study evaluated the effects of conservation practices at the edge of paired commercial cotton fields located in northeast Arkansas by measuring nutrient and sediment loads per runoff event. The cover crop was seeded after the cotton crop was harvested and terminated one or two weeks before cotton was planted, and the duration was considered the cover crop period. Similarly, cotton crop plantation to harvest was considered the cotton growing season. During the cover crop season, the runoff events were solely due to rainfall (N = 66), and during the growing season, the runoff was due to irrigation (N = 27) or rainfall (N = 28), despite missing a few common runoff events due to equipment malfunctioning or inclement weather events, evidence of the challenges of remotely located, and field-scale research. Using common per-event data from five years of cover crop treatment and four years of filter strip treatment, it was found that the use of a cover crop and filter strip at the edge of a commercial cotton field reduced runoff, sediment, and nutrient loads. The results showed reductions of 30%, 49%, and 43% in runoff depth, peak flow rate, and sediment load, respectively, due to cover crop treatment, and 36%, 49%, and 56% in the same parameters, respectively, due to filter strips. While the nutrient reduction by the cover crop and filter strip treatment was not always statistically significant, average loads of TP, PO4-P, TN, NO3-N, NO2-N, and NH4-N reduction were 4%, 21%, 7%, 9%, 4%, and 17% in the cover crop treatment and were 15%, 23%, 21%, 11%, -6%, and 42% in the filter strip treatment, respectively.
Implementing cover crops during fallow season and low-cost conservation practices, such as filter strips, can reduce pollution in the environment caused by sediment and nutrient runoff. Planting of cover crops quickly after harvest and before inclement weather can aid in maximizing benefits from cover crops. Likewise, stacking conservation practices such as reduced tillage, riparian buffer, crop rotation, modified nutrients and irrigation management, denitrifying bioreactor, etc., in the treatment field may amplify these positive effects. Positioning filter strips along field edges, particularly in turn rows and in front of drainage pipes, can contribute to nutrient and sediment load reduction and yield environmental benefits without compromising productive land.
Acknowledgments
The research team acknowledges the Natural Resource Conservation Service’s Mississippi Healthy Watershed Initiative for a financial contribution towards this effort. We also thank the producer who allowed the sample collection and provided us with the management practice information. The work was also partially supported by agreement number 58-6024-8-020 between the United States Department of Agriculture, Agricultural Research Service, and North Carolina Agricultural and Technical State University. All opinions expressed in this paper are those of the authors and do not necessarily reflect the policies of the USDA, ARS, or NC A&T State University.
Appendix
The PO4-P, NO3-N, NO2-N, and NH4-N graphs are included in the appendix. These nutrients were analyzed using the same runoff sample presented in the main section.
 |
| Figure A1. Water sampler and communication accessories inside a weather-resistant enclosure. |
 |
| Figure A2. Phosphate phosphorus load per event on yearly basis, as well as treatment combination during growing season. The red lines indicate mean, whereas horizontal black lines in box plot indicate the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
 |
| Figure A3. Nitrate nitrogen load per event on yearly basis, as well as treatment combination during growing season. The red lines indicate mean, whereas horizontal black lines in box plot indicate the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
 |
| Figure A4. Nitrite nitrogen load per event on yearly basis, as well as treatment combination during growing season. The red lines indicate mean, whereas horizontal black lines in box plot indicate the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
 |
| Figure A5. Ammonium nitrogen load per event on yearly basis, as well as treatment combination during growing season. The red lines indicate mean, whereas horizontal black lines in box plot indicate the 10th, 25th, median, 75th, and 90th percentile. C=Control, T= Treatment, CC= Cover crop, FS = filter strip. |
References
Abu-Zreig, M. (2001). Factors affecting sediment trapping in vegetated filter strips: Simulation study using VFSMOD. Hydrol. Process., 15(8), 1477-1488. https://doi.org/10.1002/hyp.220
Adviento-Borbe, M. A., Barnes, B. D., Iseyemi, O., Mann, A. M., Reba, M. L., Robertson, W. J.,... Teague, T. G. (2018). Water quality of surface runoff and lint yield in cotton under furrow irrigation in Northeast Arkansas. Sci. Total Environ., 613-614, 81-87. https://doi.org/10.1016/j.scitotenv.2017.09.020
American Public Health Association. (2005). Standard methods for the examination of water and wastewater (21st ed.). Washington, DC: APHA.
Aryal, N., Reba, M. L., Straitt, N., Teague, T. G., Bouldin, J., & Dabney, S. (2018). Impact of cover crop and season on nutrients and sediment in runoff water measured at the edge of fields in the Mississippi Delta of Arkansas. J. Soil Water Conserv., 73(1), 24-34. https://doi.org/10.2489/jswc.73.1.24
Blanco-Canqui, H. (2018). Cover crops and water quality. Agron. J., 110(5), 1633-1647. https://doi.org/10.2134/agronj2018.02.0077
Blanco-Canqui, H., Gantzer, C. J., & Anderson, S. H. (2006). Performance of grass barriers and filter strips under interrill and concentrated flow. J. Environ. Qual., 35(6), 1969-1974. https://doi.org/10.2134/jeq2006.0073
Blanco-Canqui, H., Mikha, M. M., Presley, D. R., & Claassen, M. M. (2011). Addition of cover crops enhances no-till potential for improving soil physical properties. Soil Sci. Soc. Am. J., 75(4), 1471-1482. https://doi.org/10.2136/sssaj2010.0430
Bortolozo, F. R., Favaretto, N., Dieckow, J., de Moraes, A., Vezzani, F. M., & da Silva, É. D. (2015). Water, sediment and nutrient retention in native vegetative filter strips of Southern Brazil. Int. J. Plant Soil Sci., 4(5), 426-436. https://doi.org/10.9734/IJPSS/2015/13788
Centner, T. J., Houston, J. E., Keeler, A. G., & Fuchs, C. (1999). The adoption of best management practices to reduce agricultural water contamination. Limnologica, 29(3), 366-373. https://doi.org/10.1016/S0075-9511(99)80029-2
Cheng, Q., Tang, C.-S., Xu, D., Zeng, H., & Shi, B. (2021). Water infiltration in a cracked soil considering effect of drying-wetting cycles. J. Hydrol., 593, 125640. https://doi.org/10.1016/j.jhydrol.2020.125640
Christianson, R., Fox, J., Law, N., & Wong, C. (2021). Effectiveness of cover crops for water pollutant reduction from agricultural areas. Trans. ASABE, 64(3), 1007-1017. https://doi.org/10.13031/trans.14028
Dabney, S. M. (1998). Cover crop impacts on watershed hydrology. J. Soil Water Conserv., 53(3), 207-213. Retrieved from https://www.jswconline.org/content/53/3/207.short
Dabney, S. M., Delgado, J. A., & Reeves, D. W. (2001). Using winter cover crops to improve soil and water quality. Commun. Soil Sci. Plant Anal., 32(7-8), 1221-1250. https://doi.org/10.1081/CSS-100104110
Defersha, M. B., Quraishi, S., & Melesse, A. (2011). The effect of slope steepness and antecedent moisture content on interrill erosion, runoff and sediment size distribution in the highlands of Ethiopia. Hydrol. Earth Syst. Sci., 15(7), 2367-2375. https://doi.org/10.5194/hess-15-2367-2011
Deletic, A. (2000). Sediment behaviour in overland flow over grassed areas. PhD diss. United Kingdom: University of Aberdeen. Retrieved from https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.327504
Dendy, F. E. (1981). Sediment yield from a Mississippi delta cotton field. J. Environ. Qual., 10(4), 482-486. https://doi.org/10.2134/jeq1981.00472425001000040012x
Dillaha, T. A., Sherrard, J. H., Lee, D., Mostaghimi, S., & Shanholtz, V. O. (1988). Evaluation of vegetative filter strips as a best management practice for feed lots. J. Water Pollut. Control Fed., 60(7), 1231-1238. Retrieved from https://www.jstor.org/stable/25043629
Doane, T. A., & Horwáth, W. R. (2003). Spectrophotometric determination of nitrate with a single reagent. Anal. Lett., 36(12), 2713-2722. https://doi.org/10.1081/AL-120024647
Dodds, W. K., Bouska, W. W., Eitzmann, J. L., Pilger, T. J., Pitts, K. L., Riley, A. J.,... Thornbrugh, D. J. (2009). Eutrophication of U.S. freshwaters: Analysis of potential economic damages. Environ. Sci. Technol., 43(1), 12-19. https://doi.org/10.1021/es801217q
Gavric, S., Leonhardt, G., Marsalek, J., & Viklander, M. (2019). Processes improving urban stormwater quality in grass swales and filter strips: A review of research findings. Sci. Total Environ., 669, 431-447. https://doi.org/10.1016/j.scitotenv.2019.03.072
Goel, P. K., Rudra, R. P., Khan, J., Gharabaghi, B., Das, S., & Gupta, N. (2004). Pollutants removal by vegetative filter strips planted with different grasses. Paper no. 042177. Proc. 2004 ASAE Annual Meeting. St. Joseph, MI: ASABE. https://doi.org/10.13031/2013.17669
He, S., Chen, W., Wang, D., Chen, X., Qi, Y., Zhao, P.,... Jamali, A. A. (2023). Experimental investigation of the effects of shrub filter strips on debris flow trapping and interception. Int. J. Sediment Res., 38(2), 265-278. https://doi.org/10.1016/j.ijsrc.2022.09.005
Hernandez-Santana, V., Zhou, X., Helmers, M. J., Asbjornsen, H., Kolka, R., & Tomer, M. (2013). Native prairie filter strips reduce runoff from hillslopes under annual row-crop systems in Iowa, USA. J. Hydrol., 477, 94-103. https://doi.org/10.1016/j.jhydrol.2012.11.013
Kaspar, T. C., & Singer, J. W. (2011). The use of cover crops to manage soil. In J. L. Hatfield, & T. J. Sausar (Eds.), Soil management: Building a stable base for agriculture (pp. 321-337). ASA, CSSA, SSSA. https://doi.org/10.2136/2011.soilmanagement.c21
Kassel, K. (2023). State fact sheets: Arkansas. USDA-ERS. Retrieved from https://data.ers.usda.gov/reports.aspx?StateFIPS=05&StateName=Arkansas&ID=17854
Kaushal, S. S., Groffman, P. M., Band, L. E., Elliott, E. M., Shields, C. A., & Kendall, C. (2011). Tracking nonpoint source nitrogen pollution in human-impacted watersheds. Environ. Sci. Technol., 45(19), 8225-8232. https://doi.org/10.1021/es200779e
Kim, S., Dale, B. E., Martinez-Feria, R., Basso, B., Thelen, K., Maravelias, C. T.,... Robertson, G. P. (2023). Global warming intensity of biofuel derived from switchgrass grown on marginal land in Michigan. GCB Bioenergy, 15(3), 319-331. https://doi.org/10.1111/gcbb.13024
Korucu, T., Shipitalo, M. J., & Kaspar, T. C. (2018). Rye cover crop increases earthworm populations and reduces losses of broadcast, fall-applied, fertilizers in surface runoff. Soil Tillage Res., 180, 99-106. https://doi.org/10.1016/j.still.2018.03.004
Lambrechts, T., François, S., Lutts, S., Muñoz-Carpena, R., & Bielders, C. L. (2014). Impact of plant growth and morphology and of sediment concentration on sediment retention efficiency of vegetative filter strips: Flume experiments and VFSMOD modeling. J. Hydrol., 511, 800-810. https://doi.org/10.1016/j.jhydrol.2014.02.030
Lee, K. H., Isenhart, T. M., Schultz, R. C., & Mickelson, S. K. (1998). Nutrient and sediment removal by switchgrass and cool-season grass filter strips in Central Iowa, USA. Agrofor. Syst., 44(2), 121-132. https://doi.org/10.1023/A:1006201302242
Lee, K.-H., Isenhart, T. M., Schultz, R. C., & Mickelson, S. K. (2000). Multispecies riparian buffers trap sediment and nutrients during rainfall simulations. J. Environ. Qual., 29(4), 1200-1205. https://doi.org/10.2134/jeq2000.00472425002900040025x
Lee, S., Yeo, I.-Y., Sadeghi, A. M., McCarty, G. W., Hively, W. D., & Lang, M. W. (2016). Impacts of watershed characteristics and crop rotations on winter cover crop nitrate-nitrogen uptake capacity within agricultural watersheds in the Chesapeake Bay region. PLoS One, 11(6), e0157637. https://doi.org/10.1371/journal.pone.0157637
Marttila, H., & Kløve, B. (2009). Retention of sediment and nutrient loads with peak runoff control. J. Irrig. Drain. Eng., 135(2), 210-216. https://doi.org/10.1061/(ASCE)0733-9437(2009)135:2(210)
Meisinger, J. J., Hargrove, W. L., Mikkelsen, R. L., Williams, J. R., & Benson, V. W. (1991). Groundwater impacts - Effects of cover crops on groundwater quality. In W. L. Hargrov (Ed.), Cover crops for clean water (pp. 57-68). Ankeny, IA: Soil and Water Conservation Society. Retrieved from https://www.swcs.org/static/media/cms/CCCW4ground_D30E9A7F53D56.pdf
Mitsch, W. J., Day, J. W., Gilliam, J. W., Groffman, P. M., Hey, D. L., Randall, G. W., & Wang, N. (2001). Reducing nitrogen Loading to the Gulf of Mexico from the Mississippi River Basin: Strategies to counter a persistent ecological problem. Bioscience, 51(5), 373-388. https://doi.org/10.1641/0006-3568(2001)051[0373:Rnlttg]2.0.Co;2
Odhiambo, J. J., Temple, W. D., & Bomke, A. A. (2012). Managing cover crops for conservation purposes in the Fraser River Delta, British Columbia. In F. R. Marin (Ed.), Crop management - cases and tools for higher yield and sustainability (p. Ch. 3). Rijeka: IntechOpen. https://doi.org/10.5772/33182
Prosser, R. S., Hoekstra, P. F., Gene, S., Truman, C., White, M., & Hanson, M. L. (2020). A review of the effectiveness of vegetated buffers to mitigate pesticide and nutrient transport into surface waters from agricultural areas. J. Environ. Manag., 261, 110210. https://doi.org/10.1016/j.jenvman.2020.110210
Rahman, A., Rahman, S., & Borhan, M. S. (2013). Performance evaluation of three vegetative filter strip designs for controlling feedlot runoff pollution. J. Civ. Environ. Eng., 3(1). https://doi.org/10.4172/2165-784X.1000124
Ramesh, R., Kalin, L., Hantush, M., & Chaudhary, A. (2021). A secondary assessment of sediment trapping effectiveness by vegetated buffers. Ecol. Eng., 159, 106094. https://doi.org/10.1016/j.ecoleng.2020.106094
Reba, M. L., Aryal, N., Teague, T. G., & Massey, J. H. (2020). Initial findings from agricultural water quality monitoring at the edge-of-field in Arkansas. J. Soil Water Conserv., 75(3), 291-303. https://doi.org/10.2489/jswc.75.3.291
Reeves, D. W. (2018). Cover crops and rotations. In J. L. Hatfield, & B. A. Stewart (Eds.), Advance in Soil Science: Crops residue management (1st ed., pp. 125-172). Bocaa Raton, FL: Lewis Publishers, CRC Press. https://doi.org/10.1201/9781351071246
Robertson, D. M., & Saad, D. A. (2021). Nitrogen and phosphorus sources and delivery from the Mississippi/Atchafalaya river basin: An update using 2012 SPARROW models. JAWRA Journal of the American Water Resources Association, 57(3), 406-429. https://doi.org/10.1111/1752-1688.12905
Sanderson, M. A., Schmer, M., Owens, V., Keyser, P., & Elbersen, W. (2012). Crop management of switchgrass. In A. Monti (Ed.), Switchgrass: A valuable biomass crop for energy (Vol. 1, pp. 87-112). London: Springer. https://doi.org/10.1007/978-1-4471-2903-5_4
Sharpley, A. N., & Smith, S. J. (1991). Surface water impacts - Effects of cover crops on surface water quality. In W. L. Hargrove (Ed.), Cover crops for clean water (pp. 41-49). Retrieved from https://www.swcs.org/static/media/cms/CCCW3surface_79CEC411D2D30.pdf
Sheng, L., Zhang, Z., Xia, J., Liang, Z., Yang, J., & Chen, X.-a. (2021). Impact of grass traits on the transport path and retention efficiency of nitrate nitrogen in vegetation filter strips. Agric. Water Manag., 253, 106931. https://doi.org/10.1016/j.agwat.2021.106931
Teague, T. G., & Reba, M. L. (2014). Final irrigation timing 2013 in Northeast Arkansas Cotton. In D. M. Oosterhuis (Ed.), Summaries of Arkansas cotton research 2013 (p. 40). Fayetteville, AR: Arkansas Agricultural Experiment Station, University of Arkansas System, Division of Agriculture. Retrieved from https://scholarworks.uark.edu/aaesser/36
Técher, D., & Berthier, E. (2023). Supporting evidences for vegetation-enhanced stormwater infiltration in bioretention systems: A comprehensive review. Environ. Sci. Pollut. Res., 30(8), 19705-19724. https://doi.org/10.1007/s11356-023-25333-w
Thrash, J. C., Seitz, K. W., Baker, B. J., Temperton, B., Gillies, L. E., Rabalais, N. N.,... Mason, O. U. (2017). Metabolic roles of uncultivated bacterioplankton lineages in the northern Gulf of Mexico “dead zone”. mBio, 8(5). https://doi.org/10.1128/mbio.01017-17
US-EPA. (2017). National water quality inventory: Report to congress. EPA 841-R-16/011. Washington, DC: US EPA. Retrieved from https://www.epa.gov/sites/default/files/2017-12/documents/305brtc_finalowow_08302017.pdf
US-EPA. (2022). Mississippi River/Gulf of Mexico Watershed Nutrient Task Force, 2019/2021 Report to Congress, Third Report. Washington, DC: US-EPA. Retrieved from https://www.epa.gov/system/files/documents/2022-02/hypoxia_task_force_report_to_congress_2019_21_final.pdf
Van Dijk, P. M., Kwaad, F. J., & Klapwijk, M. (1996). Retention of water and sediment by grass strips. Hydrol. Process., 10(8), 1069-1080. https://doi.org/10.1002/(SICI)1099-1085(199608)10:8<1069::AID-HYP412>3.0.CO;2-4
Vannoppen, W., Vanmaercke, M., De Baets, S., & Poesen, J. (2015). A review of the mechanical effects of plant roots on concentrated flow erosion rates. Earth Sci. Rev., 150, 666-678. https://doi.org/10.1016/j.earscirev.2015.08.011
White, M. J., & Arnold, J. G. (2009). Development of a simplistic vegetative filter strip model for sediment and nutrient retention at the field scale. Hydrol. Process., 23(11), 1602-1616. https://doi.org/10.1002/hyp.7291
Winston, R. J., Powell, J. T., & Hunt, W. F. (2019). Retrofitting a grass swale with rock check dams: Hydrologic impacts. Urban Water J., 16(6), 404-411. https://doi.org/10.1080/1573062X.2018.1455881
Woodruff, N. P., Lyles, L., Siddoway, F. H., & Fryrear, D. W. (1972). How to control wind erosion - Agriculture Information Bulletin No. 354. Washington, DC: USDA-ARS. Retrieved from https://www.google.com/books/edition/How_to_Control_Wind_Erosion/5W3XefPgG5MC?hl=en&gbpv=1&dq=Woodruff,+N.+P.+(1972).+How+to+control+wind+erosion+(No.+354).+&pg=PA2&printsec=frontcover
Yuan, Y., Book, R. S., Mankin, K. R., Koropeckyj-Cox, L., Christianson, L., Messer, T., & Christianson, R. (2022). An overview of the effectiveness of agricultural conservation practices for water quality improvement. J. ASABE, 65(2), 419-426. https://doi.org/10.13031/ja.14503