Environ Manage (2007) 39:853–866 DOI 10.1007/s00267-006-0307-6
Modeling the Impacts of Farming Practices on Water Quality in the Little Miami River Basin Susanna T. Y. Tong Æ Sarawuth Naramngam
Received: 27 August 2006 / Accepted: 28 January 2007 Springer Science+Business Media, LLC 2007
Abstract Since intensive farming practices are essential to produce enough food for the increasing population, farmers have been using more inorganic fertilizers, pesticides, and herbicides. Agricultural lands are currently one of the major sources of non-point source pollution. However, by changing farming practices in terms of tillage and crop rotation, the levels of contamination can be reduced and the quality of soil and water resources can be improved. Thus, there is a need to investigate the amalgamated hydrologic effects when various tillage and crop rotation practices are operated in tandem. In this study, the Soil Water Assessment Tool (SWAT) was utilized to evaluate the individual and combined impacts of various farming practices on flow, sediment, ammonia, and total phosphorus loads in the Little Miami River basin. The model was calibrated and validated using the 1990–1994 and 1980–1984 data sets, respectively. The simulated results revealed that the SWAT model provided a good simulation performance. For those tested farming scenarios, no-tillage (NT) offered more environmental benefits than moldboard plowing (MP). Flow, sediment, ammonia, and total phosphorus under NT were lower than those under MP. In terms of crop rotation, continuous soybean and corn–soybean rotation were able to reduce sediment, ammonia, and total phosphorus loads. When the combined effects of tillage and crop rotation were examined, it was found that NT with continuous soybean or corn–soybean rotation could greatly restrain the loss of sediments and nutrients to receiving waters. Since corn–soybean rotation
S. T. Y. Tong (&) S. Naramngam Department of Geography, University of Cincinnati, Cincinnati, Ohio 45221-0131, USA e-mail:
[email protected]
provides higher economic revenue, a combination of NT and corn–soybean rotation can be a viable system for successful farming. Keywords Tillage practice Crop rotation Water flow Water quality SWAT BASINS
Introduction In the past few decades, the rapid increase in population has led to an increase in intensive agricultural management (Chinitz 1991) with fertilization, continuous cropping, and conventional tillage [e.g., moldboard plowing (MP)]. These systems are convenient and productive, and they are needed to produce a higher yield per area to support the growing population (Gajri and others 2002). However, they often contribute to environmental deterioration, degrading soil and water resources (Burt 2001). Indeed, agricultural areas have been a major source of nonpoint source pollutants [U.S. Department of Agriculture, Natural Resources Conservation Service (USDA-NRCS) 1992]. In order to reduce soil loss and improve water quality, some scientists have suggested the use of conservation tillage and crop rotation (National Research Council 1989). In most farms, tillage is often practiced as a first step in the preparation for a soil bed suitable for seed germination and seedling development. There are different types of tillage systems, each with a different amount of plow surfaces and crop residues. Generally, conventional tillage (e.g., MP) involves plowing the entire soil surface to a depth of 10–15 centimeters and leaving less than 15% of the crop remains to cover the soil surface after planting. On the other hand, conservation tillage [e.g., no-tillage (NT)] leaves a much higher amount of crop residues (almost
123
854
100%) on the field and disturbs only a very small area of the soil surface (<10%). There are several advantages of conservation tillage over conventional tillage, including a reduction of operation time, energy, labor, and pesticide cost; an improvement in long-term productivity; and an increase in profitability for producers [U.S. Department of Agriculture, Economic Research Service (USDA-ERS) 1997; Uri 1999a; Power 1987; Miglierina and others 2000]. In addition, conservation tillage offers numerous environmental benefits. It can reduce surface runoff and mitigate soil erosion (Uri 1998; Tebrugge and During 1999; Smith and others 2001). According to Uri (1999a), the changes from conventional to conservation tillage in the United State from 1982 to 1997 have resulted in total soil erosion reduction from 8.0 to 5.2 tons/year. In addition, conservation tillage decreases the loss of nutrients, enhances water quality, as well as improves soil properties and soil moisture (Uri and others 1998; Hussain and others 1999). The better environment is attributed to the increased organic matter on the soil surface (Takken and others 2001; Wolf and Snyder 2003). With the accumulation of plant residues on the soil surface, there will be a higher microbial biomass and a lower soil pH. Consequently, there will be more nutrients in the soil layer, which will be more readily available to crops (Franzluebbers and others 1995). Nevertheless, some disadvantages may occur if conservation tillage is applied to unsuitable places or conditions, such as poorly drained soils or cool regions. Possible drawbacks of conservation tillage include greater pest problems (insects, diseases, and rodents) that in turn will result in a greater use and a higher cost of pesticides, greater soil compaction, and lower crop production (Dick and others 1986a, b). In some circumstances, conservation tillage may restrict nitrogen availability to crops. This is because the residues in the soil layers can immobilize some nitrogen, reducing nitrogen recovery in NT crops (Mengel and others 1982; Rice and Smith 1984). However, this phenomenon is often temporary. Over a long-term period, the large accumulated nitrogen pool in soil layers under NT may compensate for a reduced immobilization in the first few years. Another concern is that if large amounts of fertilizers and pesticides are applied in conservation tillage, groundwater may be contaminated (Smith 1992); however, many researchers argue that when the tillage systems are properly applied, the overall surface water quality can outweigh the potential adverse effects on groundwater quality (Baker and others 1996). They contend that a good and careful management system can minimize the disadvantages or potential negative effects of conservation tillage (Dick and others 1991; Sims and others 1994; Stonehouse 1999; Clancy and others 1993; Casady and Massey 2000). Crop rotation is a practice in which subsistent crops (e.g., corn and wheat) are rotated with other crops (e.g.,
123
Environ Manage (2007) 39:853–866
legumes). The leguminous crops can enrich the nutrient contents in the soil because they harbor some symbiotic nitrogen-fixing bacteria. Even if other nonleguminous crops are used as rotation crops, many can help in extracting nutrients from deep soil and making them available for the subsequent shallow-rooted crops. Hence, crop rotation has been known to improve soil properties (the soil organic matter, fertility, soil moisture, and rooting depth) and increase long-term crop yields (Uri 1999a; Power 1987). Furthermore, it can reduce soil loss and control insects and diseases. By lessening the needs for fertilization and pesticide, there will be lower chemical accumulations and better soil and water quality [Larney and Lindwall 1995; U.S. Department of Agriculture, Economic Research Service (USDA-ERS) 1997; Miglierina and others 2000]. However, inappropriate rotation may reduce crop yields and net farm returns when higher profitability crops are replaced by lower profitability crops. A good deal of research has been conducted on the impacts of tillage and crop rotation. Most studies investigate the monetary and economic benefits. Others that examine the hydrologic and water quality effects are mainly based on field experimental studies. Results from these projects are highly accurate, but the studies are usually very time-consuming, labor-intensive, and expensive, and they are often performed in a small area, such as at a plot scale. Quantitative information on the combined hydrologic benefits of different tillage systems and rotation practices at a watershed or a subwatershed scale is still lacking. Yet it is necessary to quantify the environmental consequences when various tillage and crop rotation systems are practiced together at such a scale. This knowledge may be useful to farmers and government agencies, enabling them to make informed decisions. In addition, it can facilitate resource managers in deriving efficient mitigation measures to ameliorate water contamination problems in agricultural watersheds. The Little Miami River (LMR) was chosen as a case study for examining the impacts of farming practices on the water quantity and quality in the receiving waters. The LMR was selected because it used to be a predominantly agricultural watershed but is now undergoing rapid land use and management changes. Many forested lands have been converted to urban areas and farmlands. Some farms have been urbanized. The rest of the farms are mostly under intensive agricultural management, but there is a trend of increasing conservative tillage and crop rotation. Moreover, the LMR is one of the most biological diverse streams in Ohio and is home to 113 fish species, some of which are rare and endangered species (Harrington 1999). Unfortunately, the water quality is deteriorating. Ohio Environmental Protection Agency (EPA) (2000) found that the LMR system is under stress. There is widespread
Environ Manage (2007) 39:853–866
Escherichia coli and Fecal coliform bacteria. An elevated level of phosphorus and other nutrients was also found (Ohio EPA 2000). The presence of excessive nutrients had led to eutrophication and algae bloom, a change in fish composition, and an increase in the number of fish anomalies. Janosy (2003) reported that abnormally high levels of arsenic, cadmium, copper, mercury, selenium, and zinc were detected in fish tissue samples collected in the river. Atrazine pollutants were also found in the East Fork Little Miami River, a tributary of the LMR (Miller 2003). Many research projects have been conducted in the LMR and its tributaries to examine the water quality problems. Tong (1990) studied the effects of Milford urbanization within the LMR and reported that watershed urbanization had caused poorer water quality. Wang (2001) noted that significantly lower water quality was observed downstream from the urban areas in the LMR. Tong and Chen (2002) postulated that an increase in surface runoff and in-stream nutrient levels was related to the changes from forests to agricultural lands and impervious urban areas. These results suggest that the cause of water quality degradation is related to the changes in land use and land management in the watershed. Although there are studies on the impacts of urbanization and land use change on the flow regime and water quality in the LMR, work on quantifying the impacts of agricultural practices is limited. To protect the water resources, more research is needed to ascertain the environmental effects of agricultural practices. This is especially important for watersheds such as the LMR because it is mainly an agricultural area. Any change in the tillage and rotation systems may have significant impacts on the water quantity and quality. This research employed the Soil and Water Assessment Tool (SWAT), a hydrologic and water quality model, in the Better Assessment Science Integrating Point and Nonpoint Sources (BASINS) GIS package (USEPA 2001) to examine the hydrologic and water quality changes in a subwatershed of the LMR basin. The main objectives were (1) to evaluate the performance of the SWAT model in the simulation of flow and water quality in the LMR and (2) to evaluate the individual and combined impacts of various farming scenarios, including two tillage systems (no-tillage and moldboard plowing) and four crop rotation methods (continuous corn, continuous soybean, corn–soybean, and soybean– corn), on water flow, sediment, ammonia, and total phosphorus loads. Our goals were to (1) suggest a feasible farming practice in terms of tillage and crop rotation that would help in reducing the adverse hydrologic and water quality effects in the study area and (2) develop a protocol for a practical tool that could be used by other researchers to model and postulate the plausible impacts of proposed land use or land management changes in a watershed.
855
Materials and Methods Study Area The LMR is located in southwestern Ohio. It covers 4550 square kilometers, spanning 11 counties. The main stream is approximately 172 kilometers long, draining directly into the Ohio River (Liu 2002). In this research, a predominately agricultural subwatershed of approximately 311 square kilometers of the LMR was selected for detail simulation of the effects of different farming practices on water quality and flow. This area, located in the upper part of the LMR, includes the Yellow Springs Creek, Goose Creek, Masses Creek, Lisbon Fork, and North Fork Little Miami (Fig 1). According to the 1990 land use map, land use types in the study area consist of 87% agriculture (73% row crop and 14% pasture), 9% forest, 4% urban, and <1% water body. Data Sets Much of the data for this study area were available from the U.S. Environmental Protection Agency (USEPA) Region 5 data set (the eight-digit Hydrologic Cataloguing Unit watershed boundaries for the LMR is 05090202), which could be extracted from the BASINS package. These data
Fig. 1 The Little Miami River Basin, Ohio
123
856
included a 1980 land use map from the U.S. Department of Agriculture (USDA), a soil map at a 30 · 30 meters cell size from STATSGO, stream characteristics and Reach File coverages from USEPA, and digital elevation models (DEMs, 30 · 30 meters resolution) from the U.S. Geological Survey (USGS). Daily climatological data (air temperature, precipitation, solar radiation, wind speed, relative humidity, etc.) for the climate station at Dayton Airport station, Ohio, were obtained from the National Climatic Data Center. This station is the closest to the LMR valley and has the longest and the most complete climatic record. The 1990 land use data were obtained from the Multi-Resolution Land Characterization Consortium. They were used as input into the SWAT hydrologic model. The flow and water quality data at the LMR near Oldtown station were acquired from the USGS (NWIS) and USEPA (STORET) Web sites. They were used in model calibration and validation. Model Selection By tracing the movement of water flow and pollutants through groundwater and surface water bodies, many hydrologic models have been developed to simulate and estimate the effects of changes in climate, land use, and management practice on flow and water quality. Among these models, only a few can simulate the impacts of farming practices. They are CREAMS (Chemicals, Runoff, and Erosion from Agricultural Management Systems) (Knisel, 1980), GLEAMS (Groundwater Loading Effects on Agricultural Management Systems) (Leonard and others 1987), EPIC (Erosion–Productivity Impact Calculator) (Williams and others 1984), SWAT (Arnold and Fohrer 2005), and MIKE SHE (Systeme Hydrologique Europeen) (Abbott and others 1986a, b). Compared to other models, SWAT is the most vigorous method because it has incorporated the algorithms from GREAMS, GLEAMS, and EPIC. Initially developed by the USDA in the early 1990s, SWAT is a continuous-time, conceptually based, watershed-scale, operational (lumped) model developed specifically to forecast the long-term impacts of watershed management practices on water flow, sediment, nutrient, and agricultural chemical yields (e.g., pesticide) in large complex watersheds with varying soil and land use conditions. MIKE SHE has comparable capacity as SWAT, but it is a commercial model. As a fully distributed, physically based model, MIKE SHE requires substantial data input and an extensive amount of parameter estimations (Heuvelmans and others 2005), yet its performance is similar to that of SWAT (El-Nasr and others 2005). Conversely, SWAT is in the public domain. SWAT version 2000 has already been integrated with GIS in the USEPA’s BASINS package through a seamless SWAT Extension Interface,
123
Environ Manage (2007) 39:853–866
and users can readily utilize the model together with many other utilities and data sets in BASINS. Technical support is available through the USEPA and USDA, their contracting companies, as well as a users’ support group on the Internet. It is therefore more convenient to use SWAT. Other advantages of SWAT are that (1) it requires little calibration and parameter adjustment since it is a semidistributed model; (2) most of the required data, such as weather, soil properties, topography, vegetation, and land management practices, are readily available from government agencies and can be extracted from the BASINS package; (3) it is computationally efficient to operate, even in large river basins; (4) it is capable of simulating the long-term impacts of management changes; (5) it simulates a large number of different physical processes, such as evapotranspiration, lateral subsurface flow, return flow from groundwater, surface runoff, nutrient cycling, erosion, and sediment yield; and (6) it has the ability to partition a watershed into a number of subwatersheds, which can be useful when different areas of the watershed are dominated by land uses or soils different enough in hydrology to impact hydrology. SWAT is used not only by the USDA but also by other federal agencies, including the USEPA, and has been applied in many watersheds in the United States (Alexander and others 2001). Studies from Cho and others (1995), Bingner and others (1997), Peterson and Hamlett (1998), Henratty and Stefan (1998), Liu (2002), Ranjan and Wurbs (2002), Grizzetti and others (2003), Chaplot and others (2004), and Tripathi and others (2005) have demonstrated that SWAT is a versatile and computationally efficient model. It is regarded as a good tool for best management practices selection (Behera and Panda 2006) and total maximum daily load analyses (Kang and others 2006; USEPA 2006a). As Di Luzio and others (2004) suggest, the SWAT model can enhance the analysis of point and nonpoint sources and watershed management decision making. In this article, SWAT in BASINS was employed to simulate the effects of agricultural management practices on water quantity and quality. BASINS BASINS is a multipurpose environmental analysis system for watershed and water quality studies developed by the USEPA. It is a GIS-based modeling package that runs on an ArcView interface. The system contains not only GIS data and spatial coverages but also many utilities (e.g., watershed delineation) and tools for assessment (e.g., data mining) and postprocessing (e.g., displaying and interpreting outputs). In addition to SWAT, three other hydrologic modeling programs are included in BASINS. The Hydrological Simulation Program–Fortran is used for
Environ Manage (2007) 39:853–866
857
assessing flow, sediment, and water quality from land use changes. The Pollution Loading Application is used to examine the annual loading of nonpoint source pollution in streams. The Enhanced Stream Water Quality Model is an in-stream water quality model that can simulate water quality in stream systems (USEPA 2001). Hence, the system provides users with the analytical capabilities to conduct sophisticated simulations. The newest version, 3.1, is free to the public and the whole system, including the user’s manual and technical notes, can be downloaded from the USEPA’s Web site (USEPA 2006b).
varies with soil, land use, slope, and changes in water content and is related to the curve number (CN): s ¼ 25:4 ðð1000=CNÞ 10Þ
where CN is the curve number for antecedent moisture condition. Its value can be obtained from the National Engineering Handbook (USDA-NRCS 2004). Erosion and sediment yield are estimated using the modified universal soil loss equation (MUSLE) (Williams 1975): Y ¼ 11:8 ðV qp Þ0:56 K C PE LS
SWAT The SWAT model is divided into several components: weather, hydrology, erosion/sedimentation, plant growth, nutrients, pesticides, agricultural management, stream routing, and pond/reservoir routing (Arnold and Fohrer 2005). It simulates the hydrology of a watershed under two phases of the hydrologic cycle: the land phase and the water (or routing) phase. The land phase simulates the amount of water, sediment, nutrient, and pesticide loadings in surface runoff as it enters the main channel of each subbasin, whereas the water phase simulates the movement of water, sediment, nutrients, and pesticides through the channel network of the watershed to the lowest pour point of the river basin. The flow simulation is based on the water balance equation: SWt ¼ SWo þ
t X
ðRi Qi ETi Pi QRi Þ
ð1Þ
i¼1
where SWt is the final soil water content in millimeters, SWo is the soil water content available for plant uptake in millimeters, t is the time in days, Ri is the amount of precipitation in millimeters, Qi is the amount of surface runoff in millimeters, ETi is the amount of evapotranspiration in millimeters, Pi is the amount of percolation in millimeters, and QRi is the amount of return flow in millimeters. Using the information on daily precipitation, maximum/minimum air temperature, solar radiation, wind speed, relative humidity, and leaf area index, SWAT models the daily average soil temperature, evaporation from soils and plants, infiltration, and flow. In SWAT, the surface runoff is estimated using the NRCS curve number method (USDA, Soil Conservation Service 1972): Q ¼ ðP 0:2 sÞ2 =ðP þ 0:8 sÞ
ð2Þ
where Q is the daily runoff in millimeters, P is the daily rainfall in millimeters, and s is a retention parameter that
ð3Þ
ð4Þ
where Y is the sediment yield, V is the surface runoff column, qp is the peak flow rate, K is the soil erodibility factor, C is the crop management factor, PE is the erosion control practice factor, and LS is the slope length and steepness factor. Based on the travel time of water flow, SWAT uses a simple flood routing method to predict the sediment yield, which is composed of the clay, silt, and fine sand materials. In terms of nutrients, SWAT tracks the movement and transformation of several forms of nitrogen, including nitrates, organic nitrogen, and ammonia. The concentration of these nitrogen species is estimated. Moreover, plant use of phosphorus, soluble phosphorus, and organic phosphorus is calculated. Compilation of the Hydrologic Model for LMR To compile the hydrologic model for LMR, the data were first extracted from BASINS and a project on LMR was created. The study area was delineated from DEM data, stream characteristics, and the Reach File version 3 (RF3) using the automatic delineation tool in BASINS. Then, the 1990 land use data were reclassified into a SWAT format and overlaid over a soil map to generate a set of parameters based on the predominant land use and soil type. Second, the weather data were compiled using the 1990–1994 weather records of the closest station, Dayton Airport station. Then, all required input files and default parameters were written and used as input map layers. Finally, the SWAT hydrological model was executed. Model Calibration and Validation After the hydrologic model for LMR was compiled and executed, it had to be properly calibrated and validated to ensure its reliability and accuracy in future simulation (Klaus and others 2005). The calibration and validation processes followed the standard procedure given in the technical notes (USEPA 2006b). The user’s support group
123
858
Environ Manage (2007) 39:853–866
Table 1 SWAT model calibration and validation results from a paired t-test (Pearson correlation R, t-test, and P value) between simulated and observed data (mean monthly) at 0.05 significant level (a = 0.05) Observed
Simulated
Difference
R
t-test
P value
Calibration (1990–1994) Water flow (m3/sec)
2.860
2.846
–0.49%
0.85
0.08
0.93
15.117
15.396
1.85%
0.89
0.24
0.81
Ammonia, NH4 (mg/L)
0.059
0.060
1.69%
0.84
0.26
0.80
Total phosphorus, P (mg/L)
0.108
0.104
–3.70%
0.89
0.38
0.70
Sediment (mg/L)
Validation (1980–1984) Water flow (m3/sec)
2.943
2.924
–0.65%
0.84
0.12
0.91
22.296
21.184
–4.99%
0.80
0.42
0.68
Ammonia, NH4 (mg/L)
0.093
0.088
–5.38%
0.76
0.40
0.69
Total phosphorus, P (mg/L)
0.136
0.142
4.41%
0.78
0.54
0.59
Sediment (mg/L)
also provided valuable recommendations regarding the choice and the adjustment of relevant input parameters used for model calibration and validation. The input parameters that were adjusted included the sensitivity of the model to the different land use/soil combinations, the curve number in the hydrology component, the effective hydraulic conductivity in the main channel alluvium, the main channel cover factor, and the percolation factors for nitrogen and phosphorus. The values used for the input parameters were all within the range specified in the user’s guide. The variables employed in the model calibration and validation included water flow, sediment load, ammonia, and total phosphorus (soluble organic and mineral phosphorus). These variables were chosen because they are the basic water quality variables and can be used to indicate if the model is performing properly. They are also continuously sampled by the USGS and USEPA and there is sufficient information for model calibration and validation. A paired t-test was used as a performance criterion to compare the observed data with the simulated outputs for both calibration and validation. It is a useful criterion because it compares each set of data on a daily basis. A t-test value of less than 1.0 and a P value greater than 0.5 indicate that the simulated and observed data are in good agreement and the model can be regarded as simulating the real-world conditions relatively accurately. If the P value is less than 0.05, it indicates a significant difference between them (Liu 2002), and the input parameters of the model must be adjusted and the model results compared again with the observed data. This procedure was repeated until the simulated data were in close agreement with the observed data. Model Calibration In this study, the hydrology was calibrated first. The simulated monthly water flow values based on 1990 land use
123
and 1990–1994 weather data were compared with the monitored values collected by the USGS and USEPA. The input parameters were adjusted by trial and error until the simulated monthly flow values were statistically close to observed data. Table 1 lists the simulation performance of SWAT. The results indicated that the model could simulate water flow over the 1990–1994 period reasonably well. A detailed analysis of the comparison revealed that the observed monthly water flows ranged from 0.283 to 9.118 m3/sec, whereas simulated flows varied from 0.465 to 9.078 m3/sec. The mean observed monthly flow was 2.860 m3/sec, whereas the mean simulated value was 2.846 m3/ sec. This suggested that the model slightly underestimated the mean water flows by approximately 0.49%. The over/ underestimations did not occur systematically, although they tended to be in summer and early fall periods (July– October). Nonetheless, it can be seen from the hydrograph in Figure 2 that the simulated flows matched observed values quite well. This was confirmed by a high correlation (0.85) between both sets of data with a t-test value of 0.08 and P value of 0.93, as shown in Table 1. In addition, the mean daily simulated and observed flows also matched (2.84 and 2.85 m3/sec, respectively) with a high degree of correlation of 0.86, t-test of 0.16, and P value of 0.88. After water flow had been properly calibrated, the next step was sediment calibration. This process is necessary because the amount of organic nutrients is directly affected by sediment load. If sediment load is not properly calibrated, it will be difficult to calibrate nutrient loads, such as total phosphorus. Unfortunately, there are no observed sediment data available for this subbasin. To solve this problem, another subbasin downstream of the LMR that included the study area and had observed sediment data, USGS 03245500 Little Miami River Milford station, was set up in the SWAT model. Based on the data set of this subbasin, the sediment load from this model was calibrated using the same input parameters. The mean simulated
Environ Manage (2007) 39:853–866
859
Fig. 2 Comparison between observed and simulated monthly flow
monthly sediment concentration was 15.396 mg/L and the observed value was 15.117 mg/L. The correlation between observed and simulated data was 0.89, with a t-test of 0.24 and P value of 0.81. Since the simulation results were within acceptable limits, the parameters for sediment calibration were applied to the model for the study area. We realized that this method of sediment calibration was not ideal, but it was probably second best. This procedure assumes that both subbasins are very similar in terms of their hydrology and sediment concentration. In this research, this assumption is supported by the fact that the slopes in both areas are rather flat (<5%). In addition, the major soil properties and soil characteristics in both subbasins are similar. For example, the soil erodibility factors (USLE_K) for both subbasins are the same. USLE_K is a key factor that can affect the sediment yield in the MUSLE, a submodel in the SWAT model (Neitsch and others 2002). Two nutrients, ammonia and total phosphorus, in surface water were included in this study. The simulated monthly ammonia concentrations ranged from 0.015 to 0.261 mg/L, and the observed values were from 0.030 to 0.350 mg/L. The mean simulated monthly ammonia was 0.060 mg/L, which was 1.69% greater than the mean observed monthly value (0.059 mg/L). The ammonia concentrations tended to be overestimated from spring to early summer (April–July) and underestimated in winter (January–March). Despite this discrepancy, the general simulation results were reasonably close to observed values because the correlation between the simulated and observed values was 0.84, with t-test of 0.26 and P value of 0.80. The simulated monthly results of total phosphorus were underestimated by 3.70% compared to the observed values. The model tended to overestimate from November to December and underestimate in summer from July to September. The simulated monthly concentrations varied from 0.009 to 0.745 mg/L, whereas the observed values
ranged from 0.030 to 1.175 mg/L. Overall, the mean monthly estimation (0.104 mg/L) matched the mean observation (0.108 mg/L) acceptably well, with a correlation of 0.89, t-test of 0.38, and P value of 0.70. Since the t-test and P value between the simulated and observed data of every variable were reasonably close (Table 1), it was considered that the LMR model was properly calibrated. Model Validation The next step was validation. This step is important because it ascertains that the calibrated model is capable of portraying realistic conditions under different environments. In the validation process, the values of all input parameters from the calibration process were maintained the same, but a different data set consisting of the 1980 land use data and the 1980–1984 weather data were used. Without modifying any other input parameters, the model was run. The simulated results were again compared to observed data using the same performance criterion as in the calibration process. If the results are unacceptable, then the model has to be recalibrated using 1990 land use and 1990–1994 weather data and validated using 1980 land use and 1980–1984 weather data until the criterion is met. If the results are satisfactory, then the model is expected to produce reasonable results and can be used to forecast future events. This procedure was repeated until the simulated data were in close agreement with the observed data in both the calibration and validation runs (McCuen and Snyder 1986). In this study, the validation results were generally acceptable. However, the differences between the validation and observation values were greater than those in the calibrations, varying from –5.38% (ammonia) to –4.99% (sediment), –0.65 (water flow), and 4.41% (total phosphorus). Nevertheless, the correlations between the mean monthly observed and simulated values in the validation
123
860 Table 2 Tillage and crop rotation scenarios used in the study No-tillage/continuous corn No-tillage/continuous soybean No-tillage/corn–soybean rotation No-tillage/soybean–corn rotation Moldboard plowing/continuous corn Moldboard plowing/continuous soybean Moldboard plowing/corn–soybean rotation Moldboard plowing/soybean–corn rotation
process were still high, ranging from 0.76 to 0.84, with t-test less than 1.0 (range, 0.40–0.12) and P value greater than 0.5 (range, 0.69–0.91). These results showed that the validation was still reasonable (Table 1) and the compiled model was capable of predicting the hydrologic effects of different tillage and crop management farming practices under current parameterization. Farming Practice Scenarios In Ohio, there were 3.7 million acres of NT, or 41% of all cropland. Two-thirds of Ohio’s soybean crop and 20% of the corn crop were planted with NT (Pollock 2003). Uri (1999b) reported that Ohio was one of the leading states in using conservation tillage, accounting for 30% in 1990 and 46% in 1994. The USDA also reported that the proportion of NT in Ohio was higher than that of moldboard tillage and other conventional tillage practices (USDA-ERS 1996). In 1995, the percentage for NT was approximately 39% and that for moldboard was approximately 10%. Furthermore, crop rotation was predicted to increase and continuous single crop would be replaced by crop rotations for approximately 20% of available cropland. For example, continuous corn would be changed to corn–soybean rotation and continuous wheat would be switched to wheat– sorghum rotation (Hennessy and others 1995; Williams and others 1990). These data suggest that many farmers in Ohio are gradually changing their agricultural practices. Because of this trend, it is prudent to investigate the separate and combined hydrologic impacts of different farming management systems. In this study, eight different farming scenarios were derived and their individual and concerted impacts on flow, sediment, ammonia, and total phosphorus on the receiving waters were compared. These scenarios were formulated from two tillage systems and four crop rotations (Table 2). The analyses of the effects of these farming practices were performed using the 1990 land use and 1990–1994 weather data. Other data and parameters in the SWAT model remained unchanged. The specifications
123
Environ Manage (2007) 39:853–866
for the planting operations were set as follows: (1) The mixing efficiency of tillage (fraction) and the depth of mixing caused by tillage were set to 0.05 and 25 millimeters, respectively, in NT and 0.95 and 150 millimeters, respectively in MP; (2) the four crop rotation practices were continuous corn (5 years of corn), continuous soybean (5 years of soybean), corn–soybean rotation (1 year of corn, followed by 1 year of soybean, 1 year of corn, 1 year of soybean, and then 1 year of corn), and soybean– corn rotation (1 year of soybean and then 1 year of corn, 1 year of soybean, 1 year of corn, and, finally, 1 year of soybean); (3) both tillage operations were applied on April 1 of every year; and (4) for corn, planting (i.e., the beginning of the growing season) started on May 3, and the harvest and kill operation was on October 3, whereas for soybean, planting began on April 15 and the harvest and kill operation was set on September 15.
Results and Discussion SWAT Simulation Performance The performance of SWAT was examined through the calibration and validation processes. It was found that the SWAT model could provide a good simulation performance in the study area because it offered realistic simulations of flow and water quality in both the 1990–1994 and the 1980–1984 periods, all of which were in statistical agreement with the observed values. This conclusion supported the findings by Chaplot and others (2004). They used SWAT to simulate water flow and nitrogen loads under a short-term (9-year) and a long-term (30-year) period and found that the results were accurate with the mean error of <10% on a monthly basis. Tolson and Shoemaker (2004) also found that SWAT could reasonably simulate the temporal and spatial nature of the measured flow and water quality. Estimating the Impacts of Various Farming Practice Scenarios on Water Flow and Water Quality The Impacts of Different Tillage Practices The effects of tillage practices on flow and water quality from the 1990–1994 simulation period are shown in Tables 3 and 4. Mean monthly water flows were lower under NT than MP in all crop management systems. Monthly flows under MP were higher than those under NT by approximately 2.95, 7.44, 7.94, and 3.64% for continuous corn, continuous soybean, corn–soybean, and soybean–corn planting systems, respectively. This result is
Environ Manage (2007) 39:853–866
861
Table 3 Simulated mean monthly flow and annual loads of major nonpoint source pollutants under different farming practices Continuous Continuous Corn– corn soybean soybean
Soybean– corn
Table 4 Comparison of the environmental effects (in terms of the mean monthly flow and annual loads of major nonpoint source pollutants) of moldboard plow with no-tillage. The data shown are the % change from no-tillage Continuous Continuous Corn– corn soybean soybean
Flow (m3/sec) No-tillage
2.89
2.96
2.93
2.91
Moldboard 2.97 plow Sediment (tons)
3.18
3.16
3.01
Flow
2.95
7.44
+7.94
+3.64
+58.80
+59.72
+64.04
+55.73
Ammonium-N
+0.50
+1.21
+4.23
+2.18
Total phosphorus
+66.31
+63.53
+62.77
+66.74
Sediment
No-tillage
1,674
1,262
1,282
1,647
Moldboard Plow
2,658
2,015
2,103
2,564
Soybean– corn
Ammonium-N (kg) No-tillage
4,466
4,195
4,225
4,178
Moldboard Plow
4,488
4,246
4,404
4,269
Total phosphorus (kg) No-tillage 11,219 Moldboard Plow
18,659
8,965
9,295
10,823
14,661
15,129
18,047
reasonable because other studies have also reported similar findings. Andraski and others (1985) observed that surface runoff from NT was 11–50% less than that from conventional tillage in a 4-year study. Choudhary and others (1997) indicated that surface runoff in NT was less than that in chisel plowing and MP at a ratio of 1:2.9:4.4. The decrease in runoff and stream flow is mainly due to the fact that in NT, crop residues are left to cover the soil surface. With more litter in the surface, more water is retained. In time, most of the water will be infiltrated to the lower soil horizons instead of leaving the field as surface runoff. Annual sediment loads under MP were greater than those under NT in all crop systems. Compared with NT, MP produced higher annual sediment loads, with increases of 58.80, 59.72, 64.04, and 55.73% for continuous corn, continuous soybean, corn–soybean, and soybean–corn crop rotations, respectively. Shipitalo and Edwards (1997), Choudhary and others (1997), and Seta and others (1993) reported similar results. Langdale and others (1992) also found that annual soil loss of 25 Mg/ha under MP had been reduced to near zero after applying conservation tillage. This is because under NT, crop residues are left undisturbed. The soils are covered and are protected from soil erosion. Similar to water flow and sediment load, the annual nutrient loads under MP were higher than those under NT. However, the amount of increase in annual ammonia loads was slight. Compared to NT, the annual ammonia loads under MP increased 0.50% in continuous corn, 1.21% in continuous soybean, 2.18% in soybean–corn rotation, and
4.23 % in corn–soybean rotation. Nitrogen, which is naturally occurring in soils and added by fertilization, is often leached to surface water and groundwater in large amounts. When the soil is plowed, more nitrogen will be lost. Thus, NT will help in reducing nitrogen loss, providing a higher soil nitrogen level than conventional tillage (Gajri and others 2002). The annual total phosphorus loads were notably affected by tillage systems, and MP yielded a much higher load than NT in all crop rotation practices. The annual loads increased approximately 66.31, 63.53, 62.77, and 66.74% under MP compared to those under NT for continuous corn, continuous soybean, corn–soybean, and soybean– corn crop rotations, respectively. Stonehouse (1999) reported concordant results. Other studies also found that NT soils had a higher phosphorus concentration at the soil surface than conventional tillage over both a short-term period of 5 years (Selles and others 1997) and a long-term period of 16 years (Salina and others 1997). Johnson and others (1995) stated that phosphorus pollution of groundwater was exceedingly rare because phosphorus was adsorbed to soil particles and not removed in large quantities by leaching. Thus, it was not surprising to find that in this study there was a lesser amount of phosphorus in surface water under NT than MP. The Impacts of Crop Rotation Practices The hydrologic and water quality impacts of crop rotation practices under different tillage systems from 1990 to 1994 are listed in Tables 3 and 5. Mean monthly flow among various crop rotation practices was slightly different. Continuous soybean farming had the highest flow, followed by corn–soybean rotation and soybean–corn rotation, compared to continuous corn under both NT and MP. This may be attributed to the fact that continuous corn has a larger amount of residues than continuous soybean, corn– soybean, and soybean–corn (Jones and others 1995). These residues help to decrease water surface runoff and reduce water flow in stream.
123
862
Environ Manage (2007) 39:853–866
Table 5 Comparison of the environmental effects (in terms of the mean monthly flow and annual loads of major nonpoint source pollutants) of continuous soybean, corn–soybean rotation, and soybean– corn rotation with continuous corn. The data shown are the % change from continuous corn farming Continuous soybean
Corn– soybean
Soybean– corn
Flow No-tillage
+2.31
+1.31
+0.59
Moldboard plow
+6.78
+6.21
+1.26
–24.61 –24.17
–23.39 –20.85
–1.61 –3.51
No-tillage
–6.06
–5.39
–6.45
Moldboard plow
–5.40
–1.88
–4.89
Sediment No-tillage Moldboard plow Ammonium-N
Total phosphorus No-tillage
–20.09
–17.15
–3.53
Moldboard plow
–21.43
–18.92
–3.28
The smallest amount of annual sediment loads was found in continuous soybean. Whereas corn–soybean had a slightly higher sediment load, the amounts in soybean–corn and continuous corn were much higher. This may be related to the fact that soybeans are usually planted at a higher density and their leaves and roots offer better protection from soil erosion than corn. The amount of sediments in continuous soybean, corn–soybean, and soybean–corn declined approximately 24.61, 23.39, and 1.61% under NT and 24.17, 20.85, and 3.51% under MP, respectively, compared to continuous corn. This result was similar to that of a study in Mississippi. In that study, Intarapapong and Hite (2002) reported that sediment load from corn–soybean rotation was 20.27% less than that from continuous corn. Sediment loads from other crop rotations were also less than those from other continuous crops in the same study. In Wisconsin, Peel (1998) found that when corn was rotated with other crops (barley and/or hay) instead of grown continuously, soil erosion could be reduced by more than 50%. Annual ammonia loads were slightly affected by crop rotations. Continuous corn produced the highest ammonia load in surface water, whereas continuous soybean, corn– soybean, and soybean–corn yielded a lower amount of ammonia. This result can be explained by the fact that legumes (e.g., soybean) can fix large amounts of nitrogen and increase the available soil nitrogen (Peel 1998). Since the nitrogen produced by legumes is organic in nature, it is often adhered to soil colloids and plant residues, so it is less
123
susceptible to leaching. As a result, a smaller amount of soil nitrogen will be washed away by surface runoff. Additionally, farmers may not have to apply as much inorganic nitrogen fertilizer, which is very soluble and more vulnerable to leaching. Hence, a rotation with more years of soybean can yield less ammonia in surface water than a rotation with more years of corn. This explained our findings that continuous soybean (5 years of soybean), soybean–corn (3 years of soybean), and corn–soybean (2 years of soybean) had a lower amount of ammonia. Drury and others (2001) reported similar results. In their study, soybean–corn rotation yielded 11% less nitrogen (nitrate) loss than continuous corn. Crop rotation remarkably influenced the annual total phosphorus loads in all crop practices under both NT and MP. Compared to continuous corn, corn–soybean and soybean–corn rotations reduced total phosphorus loads approximately 18 and 3%, respectively. Intarapapong and Hite (2002) showed that total phosphorus loads from corn– soybean and corn–corn–soybean decreased approximately 4.46 and 0.81% compared to continuous corn. Among the four tested crop rotation systems in this study, continuous soybean was the best practice for phosphorus reduction because total phosphorus load drastically declined by 20.45% under NT and 21.64% under MP. The Combined Impacts of Tillage and Crop Rotation Practice Close examination of Table 6 reveals two important findings. First, the effects of tillage on flow were greater than crop rotation. Second, when the combined impacts of tillage and crop rotation were considered, it was evident that MP/continuous soybean had the greatest flow and NT/ continuous corn had the least flow. Compared with NT/ continuous corn, the flow in MP/continuous soybean was 10.03% higher. The flow in MP/corn–soybean was 9.34% higher, MP/soybean–corn 4.15% higher, and MP/continuous corn 2.77% higher. Because corn has a larger amount of residues than soybeans (Jones and others 1995), a planting system with corn is more effective at reducing surface flow. Thus, among the tested crop rotation systems, water flow under corn, especially continuous corn, was lower. Between the two tillage systems, mean monthly flows in MP were higher than in NT. Since all plant residues were left on the fields in NT, the flows under NT were much lower for all crop systems. It is therefore reasonable to find that NT/continuous corn had the least flow. With regard to sediment, NT/continuous soybean yielded the smallest amount. Compared with NT/continuous corn, the annual sediment load of NT/continuous soybean was 24.61% less, followed by NT/corn–soybean (–23.42%). The highest was found in MP/continuous corn
Environ Manage (2007) 39:853–866
863
Table 6 Combined impacts of tillage and crop rotation practices on mean monthly flow and annual loads of major nonpoint source pollutants during 1990–1994a Scenario
Flow (m3/sec)
Sediment (tons)
Ammonia (kg)
Total phosphorus (kg)
NT/continuous corn
2.89
1674
4466
11,219
NT/continuous soybean
2.96 (+2.42%)
1262 (–24.61%)
4195 (–6.07%)
8,965 (–20.09%)
NT/corn–soybean
2.93 (+1.38%)
1282 (–23.42%)
4225 (–5.40%)
9,295 (–17.15%)
NT/soybean–corn
2.91 (+0.69%)
1647 (–1.61%)
4178 (–6.45%)
10,823 (–3.53%)
MP/continuous corn
2.97 (+2.77%)
2658 (+58.78%)
4488 (+0.49%)
18,659 (+66.32%)
MP/continuous soybean
3.18 (+10.03%)
2015 (+20.37%)
4246 (–4.93%)
14,661 (+30.68%)
MP/corn–soybean MP/soybean–corn
3.16 (+9.34%) 3.01 (+4.15%)
2103 (+25.63%) 2564 (+53.17%)
4404 (–1.39%) 4269 (–4.41%)
15,129 (+34.85%) 18,047 (+60.86%)
MP, moldboard plow; NT, no-tillage a
Numbers in parenthesis indicate the percentage change from NT/continuous corn farming
(+58.78%), followed by MP/soybean–corn (+53.17%). Even though corn was better at lessening flows, soybeans were better in terms of sediment reduction because they can effectively protect the soil from erosion. NT was clearly a better system than MP in erosion control because sediment loads were much lower in NT than in MP for all crop systems. In the case of ammonia, NT/soybean–corn had the lowest annual load. Compared to NT/continuous corn, it yielded 6.45% less ammonia, followed by NT/continuous soybean (–6.07%). Only MP/continuous corn provided a slightly higher ammonia load than the NT/continuous corn scenario (+0.49%). It was evident that NT was better than MP, and soybean was better than corn, in decreasing ammonia loads. Soybeans increase available soil nitrogen, but the nitrogen is kept in the soil colloids and crop residues and is less susceptible to be washed away and lost in the surface runoff. Therefore, a combination of NT and soybean planting system can reduce more ammonia leaching than a combination of MP and corn system. Annual total phosphorus load was lowest in NT/continuous soybean (–20.09% from NT/continuous corn), followed by NT/corn–soybean (–17.15%), and the highest was in MP/continuous corn (+66.32% from NT/continuous corn), followed by MP/soybean–corn (+60.86%). Again, it was apparent that NT was a better system than MP for decreasing the amount of phosphorus because the total phosphorus load was much lower in NT than in MP for all crop systems. Soybeans were better in terms of phosphorus reduction than corn. As a result, a combination of NT and soybean planting systems yielded a lower phosphorus load than a combination of MP and corn. These results show that NT offered more environmental benefits than MP because the flow, sediment, ammonia, and total phosphorus loads were lower in NT than in MP in all crop rotation systems. This may be due to the fact that the soil surfaces are left undisturbed and covered by plant
residues, decreasing runoff, soil erosion, and nutrient loads. The compositions of crop residues also play a key role in improving soil nutrients and soil properties, which in turn increase water infiltration and decrease surface runoff. Although continuous soybean resulted in the highest water flow, it was the best planting system to decrease sediment and nutrient loads, followed by corn–soybean rotation under both NT and MP.
Conclusions The simulation results from both the calibration and validation processes demonstrated that SWAT could offer realistic simulations of water quantity and quality. This shows that SWAT is a reliable practical tool for water quality modeling. With little modification, other researchers may adopt a similar protocol as used in this research to compile a SWAT-based hydrologic and water quality model to assess and predict the impacts of management changes in their watersheds. Regarding the farming practices, the results for the different tillage practices revealed that NT provided more beneficial environmental influences than MP in the LMR basin. Therefore, changing tillage practice from MP to NT will significantly improve environmental quality in the study area. In addition to tillage practices, crop rotation is also important in improving environmental quality. Among the tested crop rotation scenarios, continuous soybean was the best crop system to reduce sediment, ammonia, and phosphorus. The rotation systems (corn–soybean and soybean– corn) were the second best and they also could provide some benefits. This result is similar to that of Helmers (1986). When the combined effects of tillage and crop rotation practices were examined, it was found that NT with con-
123
864
tinuous soybean or a rotation of corn with soybean could greatly restrain the loss of sediments and nutrients to receiving waters. This suggests that the practice of NT, with perhaps some form of crop rotations, is a viable option for a successful farming system in this area. Other studies have indicated that the adverse effects of NT on poorly drained soils can be minimized or eliminated by crop rotations and careful long-term maintenance of the tillage system (Griffith and others 1988; Dick and others 1991; Stonehouse 1999; Hammel 1995). Indeed, as many researchers have pointed out, appropriate rotation of a cash crop with a legume coupled with NT can produce higher crop yields (Dick and Van Doren 1985; West and others 1996; Hill 2000). Choosing the correct farming system is a demanding process. The final decision is based heavily on net return, erosion reduction potential, or eligibility for government programs. For example, in this study, NT with continuous soybean may be the best practice to reduce sediment and nutrient loads, but it may not be the case when other variables, such as crop yields, are considered because corn usually provides a higher yield and net return than soybeans. Furthermore, planting continuous crops for a long period of time may make them more vulnerable to increased damage from pests, which may lead to a higher use of pesticides. To meet the goal of balancing environmental conservation with crop productivity and net benefits at a subwatershed scale, NT coupled with crop rotation may be a better agricultural practice. Because some farmers in Ohio are already changing to conservative tillage and crop rotation, this option may appeal to them. Nonetheless, different soil types, climate, and landscape (slope, aspect, etc.) may require different tillage and rotation planting systems. Each of these management systems has advantages and disadvantages. Therefore, farmers should consider all the benefits and drawbacks before selecting the practice that best suits their farmland. More research is needed to facilitate the decision-making process.
Acknowledgments A small part of the work reported in this article was funded by the U.S. Environmental Protection Agency. We are thankful to Paul Cocca of the U.S. Environmental Protection Agency for his helpful discussion about the use of BASINS. We also thank J. Arnold of the U.S. Department of Agriculture for his advice on the use of the SWAT model. Moreover, the invaluable comments given by three anonymous reviewers were greatly appreciated.
References Abbott MB, Bathurst JC, Cunge JA, O’Connel PE, Rasmussen J (1986a) An introduction to the European Hydrological Systems, Systeme Hydrologique Europeen, ‘‘SHE.’’ 1. History and
123
Environ Manage (2007) 39:853–866 philosophy of a physically based distributed modelling system. J Hydrol 87:45–59 Abbott MB, Bathurst JC, Cunge JA, O’Connel PE, Rasmussen J (1986b) An introduction to the European Hydrological Systems, Systeme Hydrologique Europeen, ‘‘SHE.’’ 2. Structure of a physically based distributed modelling system. J Hydrol 87:61– 77 Alexander RB, Smith RA, Schwarz GE, Preston SD, Brakebill JW, Srinivasan R, Pacheco PA (2001) In: Valigura R, Alexander R, Castro M, Meyers T, Paerl H, Stacey P, Turner RE (eds.) Atmospheric nitrogen flux from the watersheds of major estuaries of the United States: An application of the SPARROW watershed model. American Geophysical Union, Washington, DC, pp. 119–170 Andraski BJ, Mueller DH, Daniel TC (1985) Effects of tillage and rainfall simulation on water and soil losses. Soil Sci Soc Am J 49:1512–1517 Arnold JG, Fohrer N (2005) SWAT 2000: Current capabilities and research opportunities in applied watershed modelling. Hydrological Processes 19:563–572 Baker CJ, Saxton KE, Ritchie WR (1996) No-tillage seeding: Science and practice. Oxford University Press, Cambridge, UK Behera S, Panda RK (2006) Evaluation of management alternatives for an agricultural watershed in a sub-humid subtropical region using a physical process based model. Agric Ecosystems Environ 113:62–72 Bingner RL, Garbrecht J, Arnold JG, Srinivasan R (1997) Effect of watershed subdivision on simulation runoff and fine sediment yields. Trans Am Soc Agric Engineers 40:1329–1335 Burt TP (2001) Integrated management of sensitive catchment systems. Catena 42:275–290 Casady WW, Massey RE (2000) Costs and returns: Conservation tillage system and management. Cooperative Extension Service, Ohio State University, Wooster, Ohio Chaplot V, Saleh A, Jaynes DB, Arnold J (2004) Predicting water, sediment, and NO3-N loads under scenarios of land use and management practices in a flat watershed. Water Air Soil Pollution 154:271–293 Chinitz B (1991) A framework for speculating about future urban growth patterns in the US. Urban Studies 28(6):939–959 Cho SM, Jennings GD, Stallings C, Devine HA (1995) GIS-based water quality model calibration in the Delaware River Basin. ASAE Microfiche No. 952404. American Society of Agricultural Engineers, St. Joseph, Michigan Choudhary MA, Lal R, Dick WA (1997) Long-term tillage effects on runoff and soil erosion under simulated rainfall for a central Ohio soil. Soil Tillage Res 42:175–184 Clancy SA, Gardner JC, Grygiel CE, Biondini ME, Johnson GK (1993) Farming practices for a sustainable agriculture in North Dakota. Carrington Research Extension Center, Carrington, North Dakota Di Luzio M, Srinivasan R, Arnold JG (2004) A GIS-coupled hydrological model system for the watershed assessment of agricultural nonpoint and point sources of pollution. Trans GIS 8(1):113–136 Dick WA, McCoy EL, Edwards WM, Lal R (1991) Continuous application of no-tillage to Ohio soils. Agron J 83:65–73 Dick WA, Van Doren DM (1985) Continuous tillage and rotation combinations effects on corn, soybean, and oat yields. Agron J 77:459–465 Dick WA, Van Doren Jr DM, Triplett GB, Henry JE (1986a) Influence of long-term tillage and rotation combinations on crop yields and selected soil parameters: I. Results obtained for a Mollic Ochraqualf soil. Research Bulletin No. 1180. Ohio Agricultural Research and Development Center, Ohio State University, Wooster, Ohio
Environ Manage (2007) 39:853–866 Dick WA, Van Doren Jr DM, Triplett GB, Henry JE (1986b) Influence of long-term tillage and rotation combinations on crop yields and selected soil parameters: II. Results obtained for a typic Fragiudalf soil. Research Bulletin No. 1181. Ohio Agricultural Research and Development Center, Ohio State University, Wooster, Ohio Drury CF, Tan CS, Gaynor JD, Reynolds WD, Welacky TW, Oloya TO (2001) Water table management reduces tile nitrate loss in continuous corn and in a soybean–corn rotation. Scientific World J 2001(1):163–169 El-Nasr AA, Arnold JG, Feyen J, Berlamont J (2005) Modelling the hydrology of a catchment using a distributed and a semidistributed model. Hydrological Processes 19(3):573–587 Franzluebbers AJ, Hons FM, Zuberer DA (1995) Soil organic carbon, microbial biomass and mineralizable carbon and nitrogen in sorghum. Soil Sci Soc Am J 59:460–466 Gajri PR, Arora VK, Prihar SS (2002) Tillage for sustainable cropping. Food Product Press, New York Griffith DR, Kladivko EJ, Mannering JV, West TD, Parson SD (1988) Long term tillage and rotation effects on corn growth and yields on high and low organic matter poorly drained soil. Agron J 80:599–605 Grizzetti B, Bouraoui F, Granlund K, Rekolainen S, Bidoglio G (2003) Modelling diffuse emission and retention of nutrients in a Vantaanjoki watershed (Finland) using the SWAT model. Ecological Modelling 169(1):25–38 Hammel JE (1995) Long term tillage and crop rotation effects on winter wheat production in northern Idaho. Agron J 87:16–22 Harrington S (1999) Occurrence and distribution of fish species in the Great and Little Miami River Basins, Ohio and Indiana, pre1900 to 1998. U.S. Geological Survey, Water-Resources Investigations Report 99-4198 Helmers G (1986) An economic analysis of alternative cropping systems for east central Nebraska. Am J Alternative Agric 1:15– 22 Hennessy D, Babcock BA, Hayes D (1995) The budgetary and resource allocation effects of revenue assurance: Working paper 95-WP130. Center for Agriculture and Rural Development, Iowa State University, Ames, Iowa Henratty MP, Stefan HG (1998) Simulating climate change effects in a Minnesota agricultural watershed. J Environ Qual 27:1524– 1532 Heuvelmans G, Garcia Qujano JF, Muys B, Feyen J, Coppin P (2005) Modelling the water balance with SWAT as part of the land use impact evaluation in a life cycle study of CO2 emission reduction scenarios. Hydrological Processes 19:729–748 Hill P (2000) Crop response to tillage systems. Conservation tillage system and management. Cooperative Extension Service, Ohio State University, Columbus, Ohio Hussain I, Olsonb KR, Wanderb MM, Karlenc DL (1999) Adaptation of soil quality indices and application to three tillage systems in southern Illinois. Soil Tillage Res 50:237–249 Intarapapong W, Hite D (2002) Water quality impacts of conservation practices under environmental constraints: The case of the Mississippi Delta. 2002 Proceedings of a Regional Workshop SERA-IEG-30, Mississippi State University, Mississippi State, Mississippi Janosy SD (2003) Trace elements and synthetic organic compounds in streambed sediment and fish tissue in the Great and Little Miami River Basins, Ohio and Indiana, 1990–98. U.S. Geological Survey. Water Resources Investigations Report 02–4305 Johnson J, Loux M, Ropp G, Adams J (1995) Best management practices: Managing fertility in no-till. Ohio State University Extension, Columbus, Ohio Jones GA, Buehring NW, Ivy RL, Summers JD (1995) In: Kingery WL, Buehring N (eds.) Effect of soybean-corn rotation and
865 tillage on ground residue cover and canopy development. Special Bulletin 88-7, Mississippi State University, Mississippi State, Mississippi Kang MS, Park SW, Lee JJ, Yoo KH (2006) Applying SWAT for TMDL programs to a small watershed containing rice paddy fields. Agric Water Manage 79(1):72–92 Klaus E, Fohrer N, Frede HG (2005) Automatic model calibration. Hydrological Processes 19(3):651–658 Knisel WG (1980) CREAMS, a field scale model for chemicals, runoff, and erosion from agricultural management systems. Conservation Research Report No. 26. U.S. Department of Agriculture, Washington, DC Langdale GW, Mills MC, Thomas AW (1992) Use of conservation tillage to retard erosion effects of large storms. J Soil Water Conservation 47:257–260 Larney FJ, Lindwall CW (1995) Rotation and tillage effects on available soil water for winter wheat in a semi-arid environment. Soil Tillage Res 36:111–127 Leonard RA, Knisel WG, Still DA (1987) GLEAMS: Groundwater Loading Effects on Agricultural Management Systems. Trans ASAE 30(5):1403–1428 Liu AJ (2002) Employing land-use schemes as a mitigation strategy for the water quality impacts of global climate change. Ph.D. dissertation, University of Cincinnati, Cincinnati, Ohio McCuen RH, Snyder WM (1986) Hydrological modelling: Statistical methods and applications. Prentice-Hall, Englewood Cliffs, New Jersey Mengel DB, Nelson DW, Huber DM (1982) Placement of nitrogen fertilizer for no-till and conventional till corn. Agron J 74:515– 517 Miglierina AM, Iglesias JO, Landriscini MR, Galantini JA, Rosell RA (2000) The effects of crop rotation and fertilization on wheat productivity in the Pampean semiarid region of Argentina: Soil physical and chemical properties. Soil Tillage Res 53:129–135 Miller AJ (2003) Simulating agricultural contamination through the East Fork Little Miami River Watershed using the BASINS GIS package. M.S. thesis, University of Cincinnati, Cincinnati, Ohio National Research Council (1989) Alternative agriculture. National Academy Press, Washington, DC Neitsch SL, Arnold JG, Kiniry JR, Srinivasan R, Williams JR (2002) Soil and Water Assessment Tool user’s manual version 2000. Texas Water Resources Institute, College Station, Texas Ohio Environmental Protection Agency (2000) Biological and water quality study of the Little Miami River Basin, 1998. OEPA Technical Report Number MAS/1999-12-3. Ohio Environmental Protection Agency, Columbus, Ohio Peel MD (1998) Crop rotations for increased productivity. North Dakota State University Extension Service, North Dakota State University, Fargo, North Dakota Peterson JR, Hamlett JM (1998) Hydrologic calibration of the SWAT model in a watershed containing fragpan soils. J Am Water Resources Assoc 34(3):531–544 Pollock C (2003) No-till pioneers leave legacy in agricultural production. Ohio State University Extension, Columbus, Ohio Power JF (1987) Legumes: Their potential role in agricultural production. Am J Alternative Agric 2:69–73 Ranjan MS, Wurbs RA (2002) Scale-dependent soil and climate variability effects on watershed water balance of the SWAT model. J Hydrol 256(3–4):264–285 Rice CW, Smith MS (1984) Short term immobilization of fertilizer nitrogen at the surface of no-till and plowed soil. Soil Sci Soc Am J 48:295–298 Salinas JR, Matocha JE, Hons FM (1997) Long term tillage and nitrogen fertilization effects on soil properties of an alfison under dryland corn/cotton production. Soil Tillage Res 42:79–93
123
866 Selles F, Kochhann RA, Denardin JE, Zenter RP, Faganello A (1997) Distribution of phosphorus in a Brazilian oxisol under different tillage systems. Soil Tillage Res 44:23–34 Seta AK, Blevins RL, Frye WW, Barfield BJ (1993) Reducing soil erosion and agricultural losses with conservation tillage. J Environ Qual 22:661–665 Shipitalo MJ, Edwards WM (1997) Runoff and erosion control with conservation tillage and reduced-input practices on cropped watersheds. Soil Tillage Res 46:1–12 Sims GK, Buhler DD, Turco RF (1994) In: Unger PW (eds.) Residue management impacts on the environment. CRC Press, Boca Raton, Florida, pp. 77–98 Smith S (1992) Pesticide concentrations in shallow groundwater and surface runoff for land cropped to conventional and no-till soybeans. Proceeding of the 22nd Mississippi Water Resources Conferences, Mississippi State University, Mississippi State, Mississippi Smith S, Johnson RM, Pepperman AB (2001) Formulation and tillage effects on atrazine and alachlor in shallow ground water in upland corn production. Bull Environ Contamination Toxicol 67:113–121 Stonehouse DP (1999) Economic evaluation of on-farm conservation practices in the great lakes region of North America. Environmetrics 10(4):505–520 Takken I, Jetten V, Govers G, Nachtergaele J, Steegen A (2001) The effect of tillage induced roughness on runoff and erosion patterns. Geomorphology 37:1–14 Tebrugge F, During RA (1999) Reducing tillage density—A review of results from long term study in Germany. Soil Tillage Res 53:15–28 Tolson BA, Shoemaker CA (2004) Watershed modeling of the Cannonsville Basin using SWAT2000. Technical Report, School of Civil and Environmental Engineering, Cornell University, Ithaca, New York Tong STY (1990) The hydrologic effects of urban land use: A case study of the Little Miami River Basin. Landscape Urban Planning 19(1):99–105 Tong STY, Chen W (2002) Modeling the relationship between land use and surface water quality. J Environ Manage 66(4):377–393 Tripathi MP, Panda RK, Raghuwanshi NS (2005) Development of effective management plans for critical subwatersheds using SWAT model. Hydrological Processes 19(3):809–826 Uri ND (1998) The environmental consequences of the conservation tillage adoption decision in agriculture in the United States. Water Air Soil Pollution 103(1–4):9–34 Uri ND (1999a) Agriculture and the environment. Nova Science, New York Uri ND (1999b) Factors affecting the use of conservation tillage in the United States. Water Air Soil Pollution 116:621–638 Uri ND, Atwood JD, Sanabria J (1998) An evaluation of the environmental costs and benefits of conservation tillage. Environ Impact Assessment Rev 18:521–550 U.S. Department of Agriculture, Economic Research Service (1996) Summary of report: Crop residue management and tillage system
123
Environ Manage (2007) 39:853–866 trends. U.S. Department of Agriculture, Economic Research Service, Washington, DC U.S. Department of Agriculture, Economic Research Service (1997) Agricultural resources and environmental indicators. U.S. Department of Agriculture, Economic Research Service, Washington, DC U.S. Department of Agriculture, Natural Resources Conservation Service (1992) Soil bioengineering for upland slope protection and erosion reduction. Engineering field book, Part 650, Chapter 18. U.S. Department of Agriculture, Natural Resources Conservation Service, Washington, DC U.S. Department of Agriculture, Natural Resources Conservation Service (2004) National engineering handbook: Part 630 Hydrology, Chapter 10. U.S. Department of Agriculture, Natural Resources Conservation Service, Washington, DC. Available at http://www.policy.nrcs.usda.gov/media/pdf/H_210_630_10.pdf U.S. Department of Agriculture, Soil Conservation Service (1972) National engineering handbook, Hydrology, Section 4, Chapters 4–10. U.S. Department of Agriculture, Soil Conservation Service, Washington, DC U.S. Environmental Protection Agency (2001) Better Assessment Science Integrating Point and Nonpoint Sources: BASINS version 3.0 user’s manual. U.S. Environmental Protection Agency, Office of Water, Washington, DC U.S. Environmental Protection Agency (2006a) The mid-Atlantic state: Total maximum daily load. U.S. Environmental Protection Agency, Washington, DC. Available at http://www.epa.gov/ reg3wapd/tmdl/primer.htm U.S. Environmental Protection Agency (2006b) BASINS: Better Assessment Science Integrating Point and Nonpoint Sources. A powerful tool for managing watersheds. U.S. Environmental Protection Agency, Washington, DC. Available at http:// www.epa.gov/waterscience/basins Wang X (2001) Integrating water-quality management and land-use planning in a watershed context. J Environ Manage 61(1):25–36 West TD, Griffith DR, Steinhardt GC, Kladivko EJ, Parson SD (1996) Effects of tillage and rotation on agronomic performance of corn and soybean: Twenty years study on dark silty clay loam soil. J Production Agric 9:241–248 Williams JR (1975) In: Present and prospective technology for prediction sediment yield and sources. Sediment-yield prediction with universal equation using runoff energy factor. Proceedings of the sediment yield workshop, November 28–30, 1972. MS ARS-S-40. U.S. Department of Agriculture Sedimentation Lab, Oxford, pp. 244–252 Williams JR, Jones CA, Dyke PT (1984) A modeling approach to determining the relationship between erosion and soil productivity. Trans ASAE 27:129–144 Williams JR, Llewelyn RV, Barnaby GA (1990) Risk analysis of tillage alternatives with government programs. Am J Agric Econ 72:172–181 Wolf B, Snyder GH (2003) Sustainable soils: The place of organic matter in sustaining soils and their productivity. Food Product Press, New York