HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Tomer, M. D. Articles by Kramer, L. A. Search for Related Content PubMed PubMed Citation Articles by Tomer, M. D. Articles by Kramer, L. A. Agricola Articles by Tomer, M. D. Articles by Kramer, L. A. Related Collections Water Management Watershed and Landscape Processes Watershed-Scale Studies Best Management Practices Surface Hydrology

TECHNICAL REPORTS

Landscape and Watershed Processes

Agricultural Practices Influence Flow Regimes of Headwater Streams in Western Iowa

National Soil Tilth Lab., 2150 Pammel Dr., Ames, IA 50011

^* Corresponding author (tomer{at}nstl.gov)

Received for publication May 24, 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Agricultural tillage influences runoff and infiltration, but consequent effects on watershed hydrology are poorly documented. This study evaluated 25 yr (1971–1995) hydrologic records from four first-order watersheds in Iowa's loess hills. Two watersheds were under conventional tillage and two were under conservation (ridge) tillage, one of which was terraced. All four watersheds grew corn (Zea mays L.) every year. Flow-frequency statistics and autoregressive modeling were used to determine how conservation treatments influenced stream hydrology. The autoregressive modeling characterized variations in discharge, baseflow, and runoff at multi-year, annual, and shorter time scales. The ridge-tilled watershed (nonterraced) had 47% less runoff and 36% more baseflow than the conventional watershed of similar landform and slope. Recovery of baseflow after drought was quicker in the conservation watersheds, as evidenced by 365-d moving average plots, and 67% greater baseflow during the driest 2 yr. The two conventional watersheds were similar, except the steeper watershed discharged more runoff and baseflow during short (<30 d), wet periods. Significant multi-year and annual cycles occurred in all variables. Under ridge-till, seasonal (annual-cycle) variations in baseflow had greater amplitude, showing the seasonality of subsurface contaminant movement could increase under conservation practices. However, deviations from the modeled cycles of baseflow were also more persistent under conservation practices, indicating baseflow was more stable. Indeed, flow-frequency curves showed wet-weather discharge decreased and dry-weather discharge increased under conservation practices. Although mean discharge increased in the conservation watersheds, variance and skewness of daily values were smaller. Ridge tillage with or without terraces increased stream discharge but reduced its variability.

Abbreviations: PET, potential evapotranspiration • W1, W2, W3, W4, Watershed 1, Watershed 2, Watershed 3, Watershed 4

	INTRODUCTION

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
AGRICULTURAL MANAGEMENT PRACTICES can impact hydrology and water quality. For rain-fed agriculture, much of our knowledge of these impacts comes from research on the quality of waters discharged from agricultural areas. Water quality can be influenced by the extent of agricultural lands within a watershed (Miller et al., 1997; Schilling and Libra, 2000) and by the types of agricultural practices (e.g., rotation and tillage systems) being applied on the land (Bolton et al., 1970; Rejesus and Hornbaker, 1999; Sharpley et al., 2002).

Agricultural management also influences the partitioning of precipitation into runoff and infiltration (Radcliffe et al., 1988; Rhoton et al., 2002), and subsequently, the partitioning of infiltrated water into evapotranspiration, soil storage, and deep percolation (Black et al., 1981; Dao, 1993; Logsdon et al., 1999). This partitioning determines the amounts and timing of runoff and baseflow that comprise stream discharge. Yet effects of agricultural practices on stream hydrology have been documented in just a few instances. Some studies have examined how surface runoff is influenced by agriculture at the watershed scale. Moussa et al. (2002) conducted a modeling study of a small, grape (Vitus vinifera L.)–producing catchment with poorly drained soils and extensive drainage ditches in France. Effects of tillage vs. nontillage on runoff were measured at the field (vineyard) scale. Tillage increased infiltration and decreased runoff until subsequent rains decreased surface roughness through raindrop impacts. Their model was used to estimate how varying the extent of tillage could alter peak hydrograph flows. Dos Reis Castro et al. (1999) showed that, compared with conventional till, no-till reduced runoff from plots and small catchments cropped to soybean [Glycine max (L.) Merr.] and small-grains in Brazil. In eastern Germany, increases in the scale of agricultural operations and tractor sizes since 1950 have been related to increased flood severity (van der Ploeg and Schweigert, 2001). These studies concerned runoff events, but not stream baseflow or its variation. Effects of agricultural practices on baseflow regime¹ are more difficult to document with field data because baseflow is typically measured from watershed areas that contain mixed practices. While increased use of conservation tillage is one hypothesized cause of increases in baseflow discharge from Iowa streams during the last 50 yr (Schilling and Libra, 2003), it is rare to find small agricultural watersheds that are managed as a single field and are drained by a monitored perennial stream, where these effects could be clearly identified.

	SETTING AND BACKGROUND

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
This study used data from four small watersheds of the Deep Loess Research Station, located near Treynor, IA. The station has a research history dating back to the mid-1960s, with much of the research focused on erosion. The watersheds vary in size from 30 to 61 ha, and each forms the origin of a perennial, first-order stream. Soils developed in the deep loess (Ruhe, 1969) are dominantly Typic Hapludolls, Typic Udorthents, and Cumulic Hapludolls (Soil Survey Staff, 1994), and more than one-third of the area is considered highly eroded (Karlen et al., 1999). Significant research was also conducted on nutrient balances associated with corn production (e.g., Karlen et al., 1998; Logsdon et al., 1999).

These watersheds were consistently managed under a single experimental design between 1971 and 1995 (Karlen et al., 1999). A continuous corn rotation was present in all four watersheds. Watersheds 1 and 2 (W1, W2) were both under a conventional (full-width) tillage system (Karlen et al., 1999), which consisted of contour tillage using a moldboard plow until 1980, and offset disking thereafter. The conventional tillage disturbed 100% of the soil surface and retained little crop residue to protect the soil surface from erosion. Watersheds 3 and 4 (W3, W4) were under a ridge-till system, which maintains a "corrugated" surface with the crop planted on top of small ridges (100 mm high on a 900 mm spacing). Cultivation only occurs in the furrows between ridges, and herbicide is only applied in bands on the ridges (typically pre-emergent application). The practice increases crop-residue, and reduces net areas of tillage disturbance and herbicide application. Benefits of ridge till include reduced sediment losses in runoff (Ginting et al., 1998; Kramer et al., 1999) and flexibility for weed control (Klein et al., 1996). Orientation of ridges along the topographic contour encourages rainfall to infiltrate the soil. Watershed 4, while being ridge tilled, was also terraced. Terraces that had been in place since 1964 were widened in 1972. Herein, W3 and W4 are referred to as conservation watersheds, with W1 and W2 denoted as conventional.

Annual crop water-use efficiencies have been compared between W2 and W3 (Logsdon et al., 1999). Results showed greater water-use efficiencies in W3 than in W2, attributed in part to increased crop residues in W3 and their effect of decreasing soil evaporation. Annual surface runoff and baseflow volumes from these four watersheds have also been evaluated (Kramer et al., 1999). There was less surface runoff and greater baseflow from the conservation watersheds than from the conventional watersheds, due to greater infiltration of precipitation under the conservation practices in W3 and W4. However, these published results focused on differences in annual water yield and not on variation in discharge. Understanding this variation is crucial, because the most critical times for management of water resources are periods of flood and drought. Seasonal variations in discharge also influence the timing of contaminant fluxes from agricultural watersheds (Cambardella et al., 1999; Pionke et al., 1999; Tomer et al., 2003). Discharge has occurred continuously from all four of these watersheds since monitoring began, due to the slow movement of water through the deep loess deposits (Steinheimer et al., 1998; Tomer and Burkart, 2003). This provides a unique opportunity to analyze flow records from small watersheds without the limitations imposed by prolonged periods of zero discharge that typify this scale (Smakhtin, 2001).

	OBJECTIVES

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The objective of this research was to compare the discharge regimes of four first-order watersheds in western Iowa that were under differing agricultural management practices. We hypothesized that (i) variations in runoff, baseflow, and total stream discharge from these watersheds can be statistically quantified using autoregressive models; and (ii) conservation practices have effects on variations in discharge that can be identified using autoregressive models, and statistics describing discharge amounts, frequencies, and durations.

	METHODS

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Broad-crested weirs at the outlet of each watershed (Fig. 1) were monitored continuously, using floats and pen recorders from the start of monitoring, and then with pressure transducers and dataloggers since the early 1990s. Rainfall was monitored using tipping bucket rain gauges (locations shown in Fig. 1), and rainfall amounts in each watershed were determined using area weighting for each gauge based on Theissen polygons (Chow et al., 1988). Hydrograph separations of baseflow from runoff were made using semi-logarithmic plots of discharge data. Most runoff events were brief (<0.5 h), and a straight-line separation from the initial flow increase to the base of the hydrograph's falling limb was typically applied (Kramer et al., 1999). Snowmelt events were often diurnal, and were recorded as runoff.

View larger version (44K): [in this window] [in a new window]   Fig. 1. Maps of the four watersheds comprising the Deep Loess Research Station. The weather station in the southwest part of W3 included the evaporation pan, one of the rain gauges, and provided the temperature data used to calculate monthly PET.

Pan evaporation data were collected at a weather station between W3 and W4 (Fig. 1). These measurements were not made during winter (November through March or April). Therefore, potential evapotranspiration was also estimated on a monthly basis using the Hargreaves and Samani (1985) method, using temperature data from the same location.

Autoregressive modeling (Meek et al., 2001) was used to identify annual and longer-term cycles (and possible trends) in the data, in the presence of autocorrelation effects. This modeling was applied to monthly data for precipitation, surface runoff, baseflow, and total discharge. The monthly data were used to minimize problems associated with series of nil values that occurred in the daily data for precipitation and runoff. A square root transform was applied to minimize skew in the data. Analysis of daily values was also performed for baseflow and total discharge, which had no nil values. The generalized equation for an autoregressive model that includes an annual cycle, a longer-term cycle, and first- and second-order autoregressive terms for monthly data is as follows:

[1]

in which Q is the hydrologic discharge (dependent), and D and Y are date (independent) parameters (D is month of year for monthly data or day of year for daily data, and Y is the year expressed in decimal form). Coefficients fit by the model include a (the intercept), b, c, d₁, and d₂, along with D_m (the time of year when the annual cycle peaks), Y_L, and Y_m (the length and peak position of the long-term cycle, in years). The constant 12 is the length of the annual cycle in months (365.25 is substituted for analysis of daily data). Variables R_–1 and R_–2 are first and second order residuals (deviations of observations from trend for the two previous observations). Herein b and c are denoted as amplitude of the annual and longer-term cycle, respectively, while D_m and Y_m are denoted as "offset" positions of these cycles. The terms d₁ and d₂ quantify the persistence of deviations in observed data from the cyclic variations in the model. Fitting of the coefficients is performed through an iterative procedure, using a nonlinear least-squares regression method (Proc NLIN; SAS Institute, 1999). All coefficients included in the final models had 95% confidence intervals that excluded zero.

Comparisons of the autoregressive-model coefficients were made among watersheds to determine differences in flow regime. That is, coefficients for long-term and annual cycles and short-term (autoregressive) deviations were tested for differences among watersheds, using a t test that assumed unequal variances according to the standard errors of the coefficients. Differences in the flow regime were interpreted according to the three time scales represented in the model (multi-year, annual, and short-term), given a chosen level of significance. Here, "p values" less than or equal to 0.1 are reported and considered significant.

We evaluated baseflow recession in all four watersheds, using a method described by Witternburg and Sivapalan (1999). Results were used to determine if any differences between the watersheds could be due to subsurface controls on ground water storage and its release as baseflow. Recession rates were determined for all periods that had no precipitation for at least 10 d.

Along with differences in agricultural practices, there are minor differences in terrain among the watersheds that could influence results. However, the experimental design allows these effects to be evaluated. Watershed 1 and W2 were under conventional tillage, and comparison between them allows a terrain effect to be estimated. Steeper slopes occur in W1 (average 5.3% in W1 vs. 4.2% in W2), as measured by the surface-derivative method used in TAPES-G software (Gallant and Wilson, 2000) that was applied to a 7-m grid digital elevation model (see Karlen et al., 1999). Comparisons between W2 and W3 would best identify the effect of tillage system. These watersheds are similar in terms of shape and landform, and W3's average slope of 4.8% is intermediate between W1 and W2 (Kramer et al., 1999). Comparisons involving W4 are influenced by conservation practices that manipulated slopes and drainage in that watershed. The terraces in W4 reduced the length of the slopes, and the average slope of 3.8% in W4 is the least of the four watersheds. The W4 terrace system also includes surface water inlet and drop structures that deliver runoff from terrace impoundments to the outlet, and there is a small area of subsurface drainage in the valley bottom that contributes some water to runoff and baseflow. These sources of discharge in W4 were not monitored separately. The watersheds are similar in size, with W1, W2, W3, and W4 covering 30, 33, 43, and 61 ha, respectively.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Precipitation
An average of 812 mm of precipitation was received annually in these watersheds, or a total of about 20300 mm from 1971 to 1995 (Fig. 2A) . Annual precipitation, averaged across the watersheds, varied from 447 mm in 1988 to 1367 mm in 1993. Watershed 1 and W2 are about 4 km south of W3 and W4, and some differences in precipitation occurred. Watershed 3 and W4 received greater precipitation from the largest daily storm events (Fig. 2B). However, W1 and W2 cumulatively received about 400 mm (2%) more precipitation than W3 and W4 (Fig. 2A). We attribute this difference to measurement error; it largely accrued during the 1980s (Years 11–15 in Fig. 2A), and from small events. We noted about 10% more precipitation was recorded in W1 and W2 than W3 and W4, on average, when small amounts (<15 mm) of precipitation fell in all four watersheds (data not shown). But overall, there were only six dates when >3 mm precipitation fell on one pair of watersheds but none fell on the other pair (discounting occasional overnight rains recorded in all watersheds but on consecutive dates), and these events were all <8 mm.

View larger version (32K): [in this window] [in a new window]   Fig. 2. (A) Average cumulative precipitation for four watersheds from water years 1971 through 1995 (left axis). Cumulative deviations from the average are given on the right axis. Watershed 1 and W2 received more cumulative precipitation. However, (B) daily rainfall frequency curves show W3 and W4 received more precipitation from the heaviest storms.

View larger version (42K): [in this window] [in a new window]   Fig. 3. Monthly observations of (A) precipitation, (B) runoff, (C) baseflow, and (D) total discharge from W1 between 1971 and 1995 (dotted lines). The annual and multi-year cyclic variations determined through autoregressive modeling (given in tables) are also plotted (heavy lines). For baseflow and total discharge, the full model with an autoregressive term is also plotted (thin line).

View larger version (23K): [in this window] [in a new window]   Fig. 4. (A) Cumulative surface runoff and (B) runoff-frequency curves for four experimental watersheds during water years 1971 through 1995.

Baseflow
Conservation practices increased baseflow (Table 1). Cumulative baseflow among the watersheds (Fig. 5A) varied from 2837 mm (W1) to 4838 mm (W3). The ridge-tilled watershed (W3) had about 35% more baseflow than W2, the conventional watershed with similar terrain. Flow-frequency curves (Fig. 5B) show greater baseflow occurred from the conservation-till than the conventional-till watersheds essentially throughout the record. In the conventional watersheds, less baseflow during wet periods and greater drought persistence after dry periods are apparent when examining 365-d moving averages (Fig. 6) . The conservation watersheds also had about twice the amount of baseflow of the conventional watersheds during the driest 2-yr (730-d) period of the 25-yr record (Table 1), showing further that the impact of drought on baseflow was minimized under the conservation practices. Due to the larger baseflow and smaller surface runoff discharged from the conservation watersheds, the baseflow proportion of total discharge was larger in the conservation watersheds (0.80–0.87) than in the conventional-till watersheds (0.63–0.70; see Table 1).

View larger version (24K): [in this window] [in a new window]   Fig. 5. (A) Cumulative baseflow and (B) baseflow-frequency curves for four experimental watersheds during water years 1971 through 1995.

View larger version (31K): [in this window] [in a new window]   Fig. 6. Baseflow from the four watersheds, expressed as a 365-d moving average. Each date is plotted as the average baseflow during the prior 365 d, removing seasonal variation from the record. Four drought periods, indicated by upward-pointing arrows, each showed a slower recovery of baseflow from the conventionally tilled watersheds (W1, W2) than the conservation watersheds (W3, W4).

Autoregressive models for monthly baseflow (Table 5) identified annual and long-term cycles in the data, and significant autocorrelation terms. They accounted for 76 to 91% of the variation in monthly baseflow discharged, depending on the watershed. The model for W1 is plotted with observed monthly data in Fig. 3C. Significant differences in model coefficients among watersheds contrasted intercept (a) and annual-amplitude (b) values between conventional and conservation watersheds, indicating increased amounts of baseflow and greater seasonal variation in baseflow from the conservation watersheds (Table 5). With greater infiltration of rainfall under the conservation systems, it is logical that seasonal variation in rainfall would then increase the amplitude of seasonal variation in baseflow. Therefore, under conservation practices, it becomes more important to manage fertilizers and chemicals carefully to minimize risks of seasonal leaching.

Autoregressive models were also fit for daily baseflow (Table 6, Fig. 7) , which again identified cyclic variations and autocorrelation. However, 99% of the variation in daily baseflow from each of the four watersheds was accounted by these models. Significant differences in model coefficients occurred among the watersheds. In general, these differences segregated the conventional and conservation watersheds, for coefficients describing intercept (p < 0.013), amplitude of annual variation (p < 0.016), and autocorrelation terms (p < 0.001). In addition, autocorrelation terms differed between W3 and W4 (p < 0.001), while the intercept terms differed between W1 and W2 (p = 0.019). Otherwise, the daily autoregressive models showed similar daily baseflow regimes for the two conventional watersheds.

View larger version (34K): [in this window] [in a new window]   Fig. 7. Daily baseflow from the four watersheds during the 1971–1995 experimental period (thin dotted lines). Annual and long-term cycles identified by autoregressive models (solid lines) show greater intra-annual variation under conservation treatments (W3, W4). With autocorrelation terms, the models account for 99% of the variation in baseflow. The full models are not plotted because they are not distinguishable from the data at this scale.

Half-lives of baseflow were estimated using baseflow recessions during periods with at least 10 d between precipitation events. Results demonstrate the persistence of baseflow from these watersheds despite their small size. Baseflow typically declined only 12 to 14% in 10 d (median values), giving baseflow half-lives between 46 and 55 d (Table 1). No major differences in baseflow recession occurred among the watersheds, which is important because baseflow recession characterizes the relationship between ground water storage and baseflow discharge (Wittenburg and Sivapalan, 1999). Differences in hydrology among these watersheds are not due to subsurface flow characteristics. Nor are they due to major differences in landform: stream initiation points in these watersheds, identified using a terrain analysis method (Montgomery and Dietrich, 1992), are all within 150 m of stream headcuts in these watersheds (M. Burkart, personal communication, 2004).

There was no apparent seasonal pattern to the baseflow recession data. Seasonal patterns in baseflow recession have been used to estimate ground water contributions to riparian evapotranspiration (Wittenburg and Sivapalan, 1999). During dry summer periods in these watersheds, clear evidence for such contributions was rare, though not absent. On occasion, late-summer baseflow did decrease by 30% within 10 d. The recession figures omit data from 1 November to 1 April, when freeze–thaw conditions could influence dry-period flows. Other dry periods in spring and autumn were omitted because increases in baseflow occurred. These increases might result from late thaws without measurable runoff during spring, and from reduced crop–water use during autumn.

Total Discharge
Total discharge showed a smaller range in variation amongst the watersheds due to the contrasting effects of the conservation practices on surface runoff and baseflow. With baseflow comprising the majority of discharge, the conservation watersheds showed greater total discharge than the conventional-till watersheds (Fig. 8A) . But increased runoff resulted in maximum 7- and 30-d average discharges being greatest in the conventional watersheds (Table 1). The effect of conservation practices is to reduce discharge during wet periods and increase it during dry periods. This is evident from flow-frequency curves (Fig. 8B), which show more variation for the conventional watersheds (i.e., the steeper flow-frequency curves for W1 and W2 cross the flatter curves for W3 and W4 in Fig. 8B). Descriptive statistics (standard deviation, skewness, kurtosis) for total discharge (Table 7) also show discharge from the conservation watersheds was less variable, even though mean discharge was greater compared with the conventional watersheds. Minimum flows of all calculated durations were greatest in W4 (Table 1), indicating the combination of ridge-till and terracing acted to increase flows during drought periods.

View larger version (23K): [in this window] [in a new window]   Fig. 8. (A) Cumulative total discharge and (B) discharge-frequency curves for four experimental watersheds during the 1971–1995 experimental period.

The larger discharge from the conservation watersheds does not translate to decreased availability of soil water to crops. In fact, Logsdon et al. (1999) found that water use efficiencies were improved in W3 compared with W2. A possible drawback to increased infiltration and baseflow discharge under conservation tillage in Iowa's loess hills is that shallower water tables could decrease the strength of streambanks, making them more susceptible to collapse, which leads to gully expansion. Gully erosion and down-cutting of streams is a major concern for water resources management in this region (Prior, 1991). Also, increased infiltration can result in more leaching of nitrates and pesticides, which then can be carried to streams in baseflow (Schilling and Wolter, 2001).

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
This 25-yr study of four small watersheds demonstrated the influence of tillage practices on amounts and timing of runoff, baseflow, and total discharge in streams. We found conservation practices increase and stabilize stream discharge. Specific conclusions are:

1. Conservation practices increased baseflow and total discharge, but decreased runoff. The ridge-tilled watershed (without terracing) had 47% less runoff and 36% more baseflow than the conventionally tilled watershed. Conservation tillage decreased the amounts and frequency of runoff. A greater proportion of the discharge from the conservation watersheds occurred as baseflow, 80 to 87% vs. 63 to 70% from the conventional watersheds. The net result was greater stream discharge from the conservation watersheds (28–29% of precipitation) than from the conventional watersheds (22–25% of precipitation).
   
2. Differences in runoff and baseflow between the two conventional watersheds were most evident during wet periods of <30 d duration, when more runoff and baseflow was discharged from the steeper watershed.
   
3. Autoregressive models identified multi-year and annual cycles in precipitation, runoff, baseflow, discharge, and PET. Models for monthly data accounted for 37 to 40% of variation in precipitation, 11 to 16% of variation in monthly runoff, 76 to 91% of variation in baseflow, 24 to 59% of the variation in total discharge, and 97% of variation in calculated PET. Autoregressive models of daily baseflow accounted for 99% of variation at that finer time scale in all four watersheds.
   
4. Significant differences (p < 0.1) in autoregressive-model coefficients occurred between conservation and full-width tillage watersheds. These indicated:
   
5. Increased baseflow, increased total discharge, and less runoff occurred under conservation practices.
   
6. Increased amplitudes of annual-cycle variation occurred for baseflow under conservation practices, and for runoff under conventional tillage. Seasonal variation in precipitation had a greater impact on runoff under conventional tillage, and a greater impact on baseflow in the conservation watersheds. One implication for water quality is that under conservation practices, the seasonality of losses of nitrate and other leachable compounds can increase; therefore, nutrient and pest management practices should take this into account.
   
7. There was greater persistence of deviations from cyclic variations in baseflow and total discharge under conservation treatments. Baseflow has a stabilizing influence on discharge, allowing total discharge to be more efficiently described by the autoregressive model in the conservation watersheds, where baseflow was a greater proportion of total discharge.
   
8. Flow-frequency curves and descriptive statistics showed less discharge during wet weather and greater discharge during dry weather under conservation practices. Although total discharge was greater from the conservation watersheds, it was also more stable. Baseflow recovery after drought occurred more quickly in the conservation watersheds, as baseflow from ridge tillage was 67% greater during the driest 2-yr period. Therefore, risks of flood and drought may be reduced by conservation systems that increase infiltration and reduce runoff. Water quality implications include reduced sediment loads in surface runoff, and less fluvial erosion and sedimentation during floods. However, increased ground water recharge and water table levels could reduce streambank stability. During drought, increased minimum flows would have ecological benefits for streams.
   

	ACKNOWLEDGMENTS

 
A significant technical effort was required to obtain 25 years of daily hydrologic data in these four watersheds. Thanks to Howard Knox, Paul Muhs, Bob Poggensee, Ralph Spomer, and Mike Sukup, among others, for their contributions to that effort. Colin Greenan provided data management and prepared graphics for this manuscript, and David James prepared the maps shown in Fig. 1. Helpful suggestions to improve the manuscript were provided by Mike Burkart, Keith Schilling, and three anonymous reviewers.

	NOTES

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
¹ The term regime, as used here, refers generally to variations in discharge and the timing of those variations.

	REFERENCES

TOP ABSTRACT INTRODUCTION NOTES SETTING AND BACKGROUND OBJECTIVES METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Black, A.L., P.L. Brown, A.D. Halvorson, and F.H. Siddoway. 1981. Dryland cropping strategies for efficient water use to control saline seeps in the northern Great Plains, U.S.A. Agric. Water Manage. 4:295–311.
• Bolton, E.F., J.W. Aylesworth, and F.R. Hore. 1970. Nutrient losses through tile drains under three cropping systems and two fertility levels on a Brookston clay. Can. J. Soil Sci. 50:275–279.
• Brouwer, C., and M. Heibloem. 1986. Irrigation water management: Irrigation water needs. Training Manual 3. FAO, Rome.
• Cambardella, C.A., T.B. Moorman, D.B. Jaynes, J.L. Hatfield, T.B. Parkin, W.W. Simpkins, and D.L. Karlen. 1999. Water quality in Walnut Creek watershed: Nitrate-nitrogen in soils, subsurface drainage water, and shallow groundwater. J. Environ. Qual. 28:25–34.[Abstract/Free Full Text]
• Chow, V.T., D.R. Maidment, and L.W. Mays. 1988. Applied Hydrology. McGraw-Hill Publ. Co., New York.
• Dao, T.H. 1993. Tillage and winter wheat residue management effects on water infiltration and storage. Soil Sci. Soc. Am. J. 57:1586–1595.[Abstract/Free Full Text]
• Dos Reis Castro, N.M., A.V. Auzet, P. Chevallier, and J.C. Leprun. 1999. Land use change effects on runoff and erosion from plot to catchment scale on the basaltic plateau of southern Brazil. Hydrol. Processes 13:1621–1628.[CrossRef]
• Gallant, J.C., and J.P. Wilson. 2000. Primary topographic attributes. p. 51–85. In J. Wilson and J. Gallant (ed.) Terrain analysis: Principles and applications. J. Wiley & Sons, New York.
• Ginting, D., J.F. Moncrief, S.C. Gupta, and S.D. Evans. 1998. Corn yield, runoff, and sediment losses from manure and tillage systems. J. Environ. Qual. 27:1396–1402.[Abstract/Free Full Text]
• Hargreaves, G.H., and Z.A. Samani. 1985. Reference crop evapotranspiration from temperature. Appl. Eng. Agric. 1:96–99.
• Karlen, D.L., L.A. Kramer, D.E. James, D.D. Buhler, T.B. Moorman, and M.R. Burkart. 1999. Field-scale watershed evaluations on deep loess soils: I. Topography and agronomic practices. J. Soil Water Conserv. 54:693–704.
• Karlen, D.L., L.A. Kramer, and S.D. Logsdon. 1998. Field-scale nitrogen balances associated with long-term continuous corn production. Agron. J. 90:644–650.[Abstract/Free Full Text]
• Klein, R.N., G.A. Wicks, and R.G. Wilson. 1996. Ridge-till, an integrated weed management system. Weed Sci. 44:417–422.
• Kramer, L.A., M.R. Burkart, D.W. Meek, R.J. Jaquis, and D.E. James. 1999. Field-scale watershed evaluations on deep loess soils: II. Hydrologic responses to different agricultural land management systems. J. Soil Water Conserv. 54:705–710.
• Logsdon, S.D., D.L. Karlen, J.H. Prueger, and L.A. Kramer. 1999. Field-scale watershed evaluations on deep loess soils: III. Rainfall and fertilizer N use efficiencies. J. Soil Water Conserv. 54:711–716.
• Meek, D., D. Dinnes, D.B. Jaynes, C. Cambardella, T. Colvin, J. Hatfield, and D. Karlen. 2001. An autoregressive model for a paired watershed comparison. p. 223–231. In G. Milliken (ed.) Proc. 12th Appl. Stat. Agric. Conf., Manhattan, KS. 1–2 May 2000. Stat. Dep., Kansas State Univ., Manhattan, KS.
• Miller, C.V., J.M. Denis, S.W. Ator, and J.W. Brakebill. 1997. Nutrients in streams during baseflow in selected environmental settings of the Potomac River basin. J. Am. Water Resour. Assoc. 33:1155–1171.
• Montgomery, D.R., and W.E. Dietrich. 1992. Channel initiation and the problem of landscape scale. Science (Washington, DC) 255:826–830.[Abstract/Free Full Text]
• Moussa, R., M. Voltz, and P. Andrieux. 2002. Effects of the spatial organization of agricultural management on the hydrological behaviour of a farmed catchment during flood events. Hydrol. Processes 16:393–412.[CrossRef]
• Pionke, H.B., W.J. Gburek, R.R. Schnabel, A.N. Sharpley, and G.F. Elwinger. 1999. Seasonal flow, nutrient concentrations and loading patterns in stream flow draining and agricultural hill-land watershed. J. Hydrol. (Amsterdam) 220:62–73.
• Prior, J.C. 1991. Landforms of Iowa. Univ. of Iowa Press, Iowa City, IA.
• Radcliffe, D.E., E.W. Tollner, W.L. Hargrove, R.L. Clark, and M.H. Golabi. 1988. Effect of tillage practices on infiltration and soil strength of a Typic Hapludult soil after ten years. Soil Sci. Soc. Am. J. 52:798–804.[Abstract/Free Full Text]
• Rejesus, R.M., and R.H. Hornbaker. 1999. Economic and environmental evaluation of alternative pollution-reducing nitrogen management practices in central Illinois. Agric. Ecosyst. Environ. 75:41–53.[CrossRef]
• Rhoton, F.E., M.J. Shipatalo, and D.L. Lindbo. 2002. Runoff and soil loss from Midwestern and southeastern US silt loam soils as affected by tillage practice and soil organic matter content. Soil Tillage Res. 66:1–11.[CrossRef]
• Ruhe, R.V. 1969. Quaternary landscapes in Iowa. Iowa State Univ. Press, Ames, IA.
• SAS Institute. 1999. SAS user's guide. Ver. 8.2. SAS Inst., Cary, NC.
• Schilling, K.E., and R.D. Libra. 2000. The relationship of nitrate concentrations in streams to row crop land use in Iowa. J. Environ. Qual. 29:1846–1851.[Abstract/Free Full Text]
• Schilling, K.E., and R.D. Libra. 2003. Increased baseflow in Iowa over the second half of the 20th century. J. Am. Water Resour. Assoc. 39:851–860.
• Schilling, K.E., and C.F. Wolter. 2001. Contribution of base flow to nonpoint source pollution loads in an agricultural watershed. Ground Water 39:49–58.
• Sharpley, A.N., P.J.A. Kleinman, R.W. McDowell, M. Gitau, and R.B. Bryant. 2002. Modeling phosphorus transport in agricultural watersheds: Processes and possibilities. J. Soil Water Conserv. 57:425–439.
• Smakhtin, V.U. 2001. Low flow hydrology: A review. J. Hydrol. (Amsterdam) 240:147–186.
• Soil Survey Staff. 1994. Keys to soil taxonomy. 6th ed. USDA, Soil Conservation Service, Washington, DC.
• Steinheimer, T.R., K.D. Scoggin, and L.A. Kramer. 1998. Agricultural chemical movement through a field-size watershed in Iowa: Subsurface hydrology and distribution of nitrate in groundwater. Environ. Sci. Technol. 32:1039–1047.[CrossRef]
• Tomer, M.D., and M.R. Burkart. 2003. Long-term effects of nitrogen fertilizer use on ground water nitrate in two small watersheds. J. Environ. Qual. 32:2158–2171.[Abstract/Free Full Text]
• Tomer, M.D., D.W. Meek, D.B. Jaynes, and J.L. Hatfield. 2003. Evaluation of nitrate nitrogen fluxes from a tile-drained watershed in central Iowa. J. Environ. Qual. 32:642–653.[Abstract/Free Full Text]
• van der Ploeg, R.R., and P. Schweigert. 2001. Elbe river flood peaks and postwar agricultural land use in East Germany. Naturwissenschaften 88:522–525.[CrossRef][ISI][Medline]
• Wittenberg, H., and M. Sivapalan. 1999. Watershed groundwater balance estimation using streamflow recession analysis and baseflow separation. Hydrology 219:20–33.[CrossRef]

Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: 0. [Full Text]  

This article has been cited by other articles:

				A. N. Sharpley, P. J. A. Kleinman, A. L. Heathwaite, W. J. Gburek, G. J. Folmar, and J. P. Schmidt Phosphorus Loss from an Agricultural Watershed as a Function of Storm Size J. Environ. Qual., March 1, 2008; 37(2): 362 - 368. [Abstract] [Full Text] [PDF]

				M. D. Tomer, C. A. Cambardella, D. E. James, and T. B. Moorman Surface-Soil Properties and Water Contents across Two Watersheds with Contrasting Tillage Histories Soil Sci. Soc. Am. J., February 27, 2006; 70(2): 620 - 630. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Tomer, M. D. Articles by Kramer, L. A.