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) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Buda, A. R. Articles by Feyereisen, G. W. PubMed PubMed Citation Articles by Buda, A. R. Articles by Feyereisen, G. W. Agricola Articles by Buda, A. R. Articles by Feyereisen, G. W. Related Collections Water Quality Watershed and Landscape Processes Phosphorus Nutrient Management Water Pollution

TECHNICAL REPORTS

Landscape and Watershed Processes

Effects of Hydrology and Field Management on Phosphorus Transport in Surface Runoff

^a USDA-ARS, Bldg. 3702 Curtin Road, University Park, PA 16802
^b USDA-ARS, 439 Borlaug Hall, 1991 Upper Buford Circle, St. Paul, MN 55108
^c National Institute of Water & Atmospheric Research (NIWA), PO Box 8602, Christchurch, New Zealand

^* Corresponding author (Anthony.Buda{at}ars.usda.gov).

Received for publication December 5, 2008.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Phosphorus (P) losses from agricultural landscapes arise from the interaction of hydrologic, edaphic, and management factors, complicated by their spatial and temporal variability. We monitored sites along two agricultural hillslopes to assess the effects of field management and hydrology on P transfers in surface runoff at different landscape positions. Surface runoff varied by landscape position, with saturation excess runoff accounting for 19 times the volume of infiltration excess runoff at the north footslope position, but infiltration excess runoff dominated at upslope landscape positions. Runoff differed significantly between south and north footslopes, coinciding with the extent of upslope soil underlain by a fragipan. Phosphorus in runoff was predominantly in dissolved reactive form (70%), with the highest concentrations associated with upper landscape positions closest to fields serving as major sources of P. However, the largest loads of P were from the north footslope, where runoff volumes were 24 times larger than from all other sites combined. Loads of P from the north footslope appeared to be primarily chronic transfers of desorbed soil P. Although runoff from the footslope likely contributed directly to stream flow and hence to stream water quality, 27% of runoff P from the upslope sites did not connect directly with stream flow. Findings of this study will be useful for evaluating the critical source area concept and metrics such as the P-Index.

Abbreviations: DRP, dissolved reactive phosphorus • PP, particulate phosphorus • TP, total phosphorus

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
TRANSFERS of phosphorus (P) from terrestrial to aquatic ecosystems contribute to accelerated eutrophication, which is the most extensive impairment of surface waters in the USA (USEPA, 2007). Agriculture is an important source of P to surface waters (e.g., Carpenter et al., 1998; Howarth et al., 2004), and efforts to reduce P pollution from agriculture have relied largely on the critical source area concept of targeting remedial practices to fields where high concentrations of P coincide with areas in the landscape prone to surface runoff generation (Sharpley et al., 1994; Gburek et al., 1996; Pionke et al., 2000). The Phosphorus Site Assessment Index (P-Index) was developed to identify fields that are critical source areas, and variations on that Index have now been adopted by the vast majority of US states (Sharpley et al., 2003). Validation of the P-Index could be greatly enhanced by studies that explore how P transport in surface runoff is affected by the interaction of field management and hydrology at plot, field, and landscape scales.

Field management of manure and fertilizer nutrients is an important factor that influences rapid incidental (e.g., Haygarth and Jarvis, 1999; Haygarth et al., 2000) and chronic transfers of P in surface runoff. Rapid incidental transfers of P are mostly likely to occur when P applications are immediately followed by large rainfall events. Recent studies have observed rapid incidental P transfers following applications of fertilizer (Scholefield and Stone, 1995; Haygarth and Jarvis, 1997; McDowell and Catto, 2005) and manure (Misselbrook et al., 1995; Wang et al., 1996; Withers et al., 2001; Preedy et al., 2001). Chronic P transfers in surface runoff are more symptomatic of repeated applications of manure and fertilizer over the long term, which can lead to increases in soil test P and ensuing desorption of P from soil solids to runoff water (Sharpley et al., 1994). Thus, it is well established that soil P and dissolved reactive P (DRP) in surface runoff are strongly tied (see review by Vadas et al., 2005). The linkage between field management and the relative importance of rapid incidental and chronic transfers of P in surface runoff is important to establish over the long-term to improve nutrient management in agricultural watersheds.

In addition to understanding field management factors, hydrology is also important because surface runoff serves as the primary flow pathway for P transport (Sharpley et al., 1994). Identification and quantification of surface runoff is complicated by the interaction of climate, geomorphic variables, and management, which can result in heterogeneity of runoff production mechanisms over space and time (Srinivasan et al., 2002). Increasingly, variable source area hydrology is recognized in watersheds with soils that are shallow or possess a discontinuity that can perch shallow water tables (Srinivasan et al., 2002; Needelman et al., 2004; Gburek et al., 2006). In such watersheds, surface runoff may be generated by infiltration or saturation excess mechanisms (Ward, 1984).

Infiltration excess runoff occurs when the rainfall rate exceeds the soil infiltration capacity (Horton, 1933), and therefore rainfall intensity is an important factor affecting the occurrence of infiltration excess runoff. Saturation excess runoff occurs from direct precipitation onto saturated areas or from return flow of subsurface water to the land surface (Dunne and Black, 1970). Several recent studies have simulated return flow (exfiltrating soil water) in the laboratory and have shown that this flow pathway can have profound effects on P concentrations and losses in surface runoff (Zheng et al., 2004; Sánchez and Boll, 2005). Therefore, identifying runoff generation mechanisms is important from the perspective of P release and transport in surface runoff.

Given the role of geomorphic processes in the development of distinct soil properties, landscape position has been shown to be an important variable affecting runoff generation processes and therefore also may be important for tracking P losses at larger scales, such as the hillslope or watershed scale. Research in humid temperate agricultural watersheds consistently has shown that saturation excess runoff is more common in the near-stream zone (Gburek et al., 2006). In some regions where colluvial soils have led to fragipan development in near-stream zones, saturation excess runoff can be extremely important due to the influence of seasonally perched water tables (Needelman, 2002; Gburek et al., 2006). In fact, a recent study by Buda et al. (2009) showed that saturation excess runoff from a fragipan soil accounted for a substantial portion of hillslope runoff in the footslope of a small agricultural watershed, whereas upslope areas produced much smaller volumes of mostly infiltration excess runoff. The same study also highlighted runoff timing and frequency differences between upper and lower landscape positions, with the footslope positions typically producing the most frequent and largest volumes of surface runoff. As a result, the timing and frequency of runoff at a given landscape position may be important for detecting the impacts of management activities on P transport.

The vast majority of runoff P studies have focused on P transfer at the plot scale for short time periods using simulated rainfall experiments (Dougherty et al., 2004 and references therein). Fewer studies have been conducted using natural rainfall at the landscape scale. Herein, we report the results of a 2.5-yr study designed to evaluate the effects of field management and hydrology on P transfers in surface runoff at different landscape positions under different rainfall conditions. Runoff was monitored continuously to assess trends in P release at three landscape positions along a single hillslope and one landscape position at the base of another hillslope. Results from this study will be useful for evaluating the critical source area concept and metrics such as the P-Index.

	Materials and Methods

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Study Area
The study was conducted on two hillslopes within the Mattern watershed (Fig. 1 ), a small, 11-ha research watershed that has been the focus of recent investigations of runoff generation processes (Buda et al., 2009) as well as nutrient transport in surface runoff under simulated rainfall (Kleinman et al., 2006). The watershed falls within the Ridge and Valley Physiographic Province of east-central Pennsylvania and has a humid temperate climate with average annual precipitation of approximately 1060 mm based on 35 yr of data (1968–2002) collected near Klingerstown, PA. Mean monthly temperatures in the watershed range from –4°C in January to 21°C in July.

View larger version (27K): [in this window] [in a new window]   Fig. 1. Map of the Mattern Watershed location within the Chesapeake Bay (gray-shaded region in upper left panel) and the WE-38 Experimental Watershed (upper right panel) as well as the distribution of fields, soils, and runoff sampling locations within the Mattern Watershed. Values shown in parentheses represent the minimum and maximum Mehlich-3 P concentrations (mg kg–1) for each field.

During the course of the study (2002–2004), the contour-cropped watershed contained four distinctly managed fields (Fig. 1). Fields 1 and 4, located at the top and base of the hillslope, respectively, were planted with alfalfa (Medicago sativa L.), were not tilled, and did not receive nutrient applications (manure/fertilizer). Field 2 was planted with corn (Zea mays L.) in 2002, soybeans (Glycine max L.) in 2003, and oats (Avena sativa L.) in 2004. Field 3 was planted with soybeans in 2002, corn in 2003, and soybeans in 2004. Poultry litter was applied to fields 2 and 3 during the second and third year of the study (69 kg ha^–1) and to field 3 only in the third year (161 kg ha^–1). These rates provided a consistent P load to the hillslope. Additional information on field management is available in Table 1 .

At each landscape position, four pairs of 1-m wide by 2-m long runoff plots were installed along a single elevation line (Fig. 2 ). Each pair of plots consisted of a closed and an open plot, which were designed to test whether flow pathway length affected P transport in surface runoff. Closed plots were isolated on the top three sides by steel frames, which only permitted runoff collection from within the plot itself. The steel frames were 10 cm high, with the first 5 cm of the frame driven into the soil and the remaining 5 cm extending above the ground surface. Open plots did not have the solid steel frame on the upper 1-m-wide side, and therefore runoff collected from this plot could include water that originated upslope of the plot. Gutters parallel to the soil surface were inserted 5 cm into the soil at the lower end of open and closed plots to exclude direct rainfall input during runoff events. Runoff from the gutter was routed by 2-cm-diameter plastic tubes to 10.5-L containers buried below the base of each plot. For the majority of events, total runoff volume was measured volumetrically; however, flow splitters were installed in July 2004 to divert a small fraction of total runoff to an automated tipping bucket rain gauge so that runoff volumes from large events could be measured more accurately (see Buda et al., 2009). Subsamples of runoff water were collected in 900-mL plastic bottles and stored at 4°C before laboratory analysis.

View larger version (29K): [in this window] [in a new window]   Fig. 2. Layout and design of open and closed runoff plots.

Soil Sampling and Laboratory Analysis
To characterize soil P levels within the watershed, surface soils were sampled (2 cm in diameter, 5 cm deep) in late April 2001 using a 30-m grid of sampling points spread out across the watershed. Soils were air dried, sieved (2-mm), and analyzed for Mehlich-3 P by agitating 2.5 g of soil with 25 mL of Mehlich-3 solution (0.2 mol L^–1 CH₃COOH + 0.25 mol L^–1 NH₄NO₃ + 0.015 mol L^–1 NH₄F + 0.013 mol L^–1 HNO₃ + 0.001 mol L^–1 EDTA) for 5 min (Mehlich, 1984). Extract P was determined colorimetrically, using a modified method of Murphy and Riley (1962), with a spectrophotometer wavelength of 712 nm. Water-extractable soil P was measured by agitating 0.5 g of soil in 5 mL of distilled water for 1 h (soil:solution = 1:10), filtering the supernatant through a Whatman No. 1 paper filter and determining P colorimetrically.

Dissolved reactive P and total P (TP) were measured on runoff samples with and without 0.45-µm filtration, respectively. Dissolved reactive P was determined colorimetrically on filtered runoff samples. Total P was measured on unfiltered runoff water by modified semimicro-Kjeldahl procedure following Bremner (1996). Total solids were determined by gravimetric analysis after evaporating 200 mL of runoff water at 80°C.

Data Analysis
In many studies of P loss in surface runoff, it is common to report the loads as yield per unit area (e.g., kg ha^–1). When collecting surface runoff in small plots, as was done in this study, the tacit assumption must be made that the contributing area for each runoff plot is clearly identifiable and relatively constant from event to event. In the case of known variable source area hydrology landscapes, such as the Mattern watershed (see Buda et al., 2009), the contributing area for runoff can vary widely depending on antecedent moisture conditions and individual storm characteristics (e.g., rainfall amount, duration, and intensity). Therfore, we report nutrient loads (mass only) by multiplying the measured concentration times the total volume of runoff for each event.

To discriminate between the contribution of rapid incidental transfers and chronic diffuse transfers on P loads, we first identified those events for which only chronic diffuse transfers of soil P to runoff were expected to play a role. Average concentrations of TP from these events were then subtracted from TP concentrations of events for which incidental transfers of fertilizer and manure P were likely and runoff volumes from those events were used to estimate the load of TP associated with incidental transfer. By summing the TP loads attributed to rapid incidental transfers, it was possible to quantify their importance relative to those from chronic transfers at different positions in the landscape.

For statistical analysis, all load and concentration data were logarithmically (natural) transformed to comply with the assumption of Gaussian distribution and evaluated by parametric methods. One-way ANOVA was used to determine whether concentrations and loads differed due to landscape position. Paired t tests were used to assess differences due to storm characteristics (e.g., high versus low rainfall amounts, durations, and intensities) and runoff generation mechanisms (infiltration versus saturation excess). Least squares regression was used to assess relationships between individual water quality variables (e.g., sediment versus particulate P concentration in runoff). All differences reported in the text are significant at = 0.05. Statistical analyses were conducted using SAS Version 9.1 (SAS Institute, 2004).

	Results and Discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Trends in Climate and Hydrology
Over the 3-yr study period, annual precipitation in the Mattern watershed ranged from approximately the long-term average (105.6 cm based on 35 yr of data) in 2002 to 15 to 20% above that average in 2003 and 2004 (Table 2 ). All storms but one had return periods of less than 2 yr based on the methods of Aron et al. (1987) and produced rainfall of 0.2 to 6.4 cm (average, 1.96 cm) and rainfall intensities of 0.1 to 2.4 cm h^–1 (average, 0.6 cm h^–1). The only storm with a return period greater than 2 yr resulted from the remnants of Hurricane Ivan (19–20 Sept. 2004), which had a 10-yr return period. Rainfall from this storm totaled 14.7 cm over 27 h, and maximum rainfall intensity reached 2.9 cm h^–1.

Phosphorus Transfers by Landscape Position
Phosphorus Concentrations.
Phosphorus concentrations in surface runoff clearly differed by landscape position, reflecting differences in P sources, hydrology, and transport processes across the landscape. Dissolved reactive P was the dominant form of TP in runoff at all landscape positions, with the DRP:TP ratio averaging 0.70 across all sites (range, 0.56–0.86) (Table 3). Mean runoff DRP and TP concentrations at the seepage slope and transportational midslope positions were significantly greater than concentrations at the north and south footslope positions, with mean concentrations at the transportational midslope position 10 and 4 times greater than those observed at the two footslope positions, respectively (Table 3).

Differences in DRP concentrations between landscape positions point to varying sources of recently applied P and soil P within the landscape as well as to the dilution of P in runoff at the footslope sampling sites. During the study period, manure and fertilizer were applied to fields 2 and 3 only (Fig. 1; Table 1), which therefore served as the only known sources of recently applied P. Therefore, one would expect incidental transfers of applied P to more directly affect runoff at the seepage slope and transportational midslope sampling sites that were established to collect runoff from the edges of these two fields. In addition, historical differences in P applications to fields in the watershed were manifest in a roughly twofold increase in Mehlich-3 P from fields 4 to 2, resulting in substantial differences in the potential for runoff P enrichment from soil P desorption. Indeed, over the course of the study, average concentrations of DRP and Mehlich-3 P from the north hillslope sampling sites were related (Fig. 3a ), although the slope of the relationship (0.009) was closer to that reported by Vadas et al. (2005) (0.002–0.005) for a broad range of soils when events representing rapid incidental transfers of applied fertilizer and manure P were excluded from the analysis (Fig. 3b).

View larger version (17K): [in this window] [in a new window]   Fig. 3. Relationship between dissolved reactive phosphorus (DRP) concentrations in runoff water and Mehlich-3 P concentrations in soil at the three landscape positions along the north hillslope for (a) annual conditions that included both incidental and chronic P transfers and (b) events that only produced chronic P transfers. Variances in DRP concentrations over the period of the study are presented as standard error bars. Mehlich-3 P concentrations represent an average of the two soil sampling points closest to the runoff plots at each landscape position.

Phosphorus Loads.
Phosphorus loads in runoff summed across all events (2002–2004) from a particular landscape position varied by a factor of six, from 0.13 g from the seepage slope to more than 0.75 g from the north footslope (Fig. 4 ), highlighting interactions between hydrology, soils, and management on P transfers. Trends in runoff P loads contrasted with those witnessed for P concentrations by landscape position because the largest loads were from the north footslope where P concentrations were lowest but runoff volumes were greatest. Indeed, across all sites, runoff volume for an individual event explained more of the variability in TP loads over the period of the study (r² = 0.52) than did runoff TP concentration (r² = 0.05). Thus, despite its relative distance from fields with the greatest potential for P release to runoff, the north footslope position had substantially greater P loads than the other sampling sites on the north hillslope (Fig. 4). In addition, TP loads from the north footslope contrasted greatly with those from the south footslope, despite their similar position within the landscape, because the lesser runoff volumes from the south footslope sampling site were coincident with a smaller contributing area of the somewhat poorly drained Albrights soil and hence lower potential for upslope runoff generation by saturation excess. Such differences confirm the generalization that areas underlain by fragipan should be targeted as the primary areas of P loss in runoff (e.g., Needelman, 2002).

View larger version (17K): [in this window] [in a new window]   Fig. 4. Dissolved reactive, particulate, and total phosphorus loads in runoff (kg) from the seepage slope, transportational midslope, and north and south footslopes for all events, frequently occurring storm events with <2-yr return periods, and Hurricane Ivan, a 10-yr return period storm that occurred in September 2004.

Seasonal Elevation in Runoff Phosphorus Concentration
Temporal trends in P concentrations in surface runoff provided insight into the role of field management, particularly manure and fertilizer application, in P transfers (Fig. 5 ). Along the north hillslope, a seasonal pattern in P concentrations was clearly evident. At the seepage slope and transportational midslope, P concentrations in surface runoff were at their lowest levels from late December through early April, averaging 1.2 mg L^–1 DRP and 1.4 mg L^–1 TP. Beginning in late April, runoff P concentrations began to increase at both landscape positions (Fig. 5a), reaching peak levels around July for the transportational midslope (averaging 20.9 mg L^–1 DRP and 25.9 mg L^–1 TP) and from September to December for the seepage slope (averaging 10.9 mg L^–1 DRP and 11.7 mg L^–1 TP). At the north footslope, P concentrations remained at or below 1 mg L^–1 for most of the monitoring period, but a clear peak was observed each year between late June and early July (Fig. 5b). No clear seasonal pattern in P concentrations was evident at the south footslope (Fig. 5b).

View larger version (20K): [in this window] [in a new window]   Fig. 5. Total phosphorus concentrations in surface runoff over the study period at (a) seepage slope and transportational midslope and (b) north and south footslopes. Arrows indicate the approximate dates of manure and fertilizer application.

A stronger alternate hypothesis for the elevated P in surface runoff is the concentration of P due to low runoff volumes. Indeed, there is a coincidence of high P concentrations at low flows for many of the summertime events. Kleinman et al. (2006) reported a change in TP concentration of 2.5-fold during the period of a single rainfall simulation event for the same experimental hillslope, attributing the decline in TP concentration to dilution. In this study, TP concentration peaks at the seepage and transportational midslope positions ranged from 4-fold (average) to 27-fold (maximum) above the TP concentrations that we defined as chronically desorbed from soils (1.1 mg P L^–1), suggesting other factors at play beyond simple dilution/concentration effects. Thus, although the coincidence of low runoff volumes and available P sources cannot be further assessed in this study, the evidence strongly points to incidental transfers of manure and fertilizer as the principal source of high P concentrations (>1.1 mg P L^–1) in surface runoff during the summer and fall months of the year.

By identifying events relatively unaffected by incidental transfers of fertilizer and manure P (i.e., fall and winter events with low runoff P concentrations), we estimate that runoff P concentrations associated with soil P desorption to runoff were 1.2, 1.1, and 0.32 mg L^–1 for the seepage slope, transportational midslope, and north footslope, respectively. This resulted in a strong relationship in runoff DRP concentration with Mehlich-3 P (Fig. 3b) much closer to that reported by Vadas et al. (2005) from a literature review of experimental data than the relationship from all events (Fig. 3a). Using the average TP concentrations from these events, we evaluated all events over the course of the study to discriminate between the contribution of chronic transfers associated with soil P desorption and soil erosion and the contribution of incidental transfers of manure and fertilizer P. Although estimates are constrained by the complexity of interactions in source and transport factors (expanded below), incidental transfers of fertilizer and manure P are clearly more important to landscape positions close to the application point. In fact, we estimate that incidental transfers account for roughly 49 and 56% of the TP load from the seepage slope and transportational midslope, respectively, but contribute <1% of the TP load from the north footslope.

The timing of the peaks relative to P application and landscape position suggests considerable complexity in the process of P transfer beyond simple wash-off of fertilizer and manure P or concentration due to low flows. First, there was a significant lag between the time of manure and fertilizer application and the concentration peak observed at each landscape position (65–160 d). Elevated (>1.1 mg P L^–1), but not peak, TP concentrations typically occurred within 27 d of manure and fertilizer application. During the lag periods between manure/fertilizer application and TP concentration peaks in surface runoff, anywhere from six (north footslope, 2004) to approximately 34 (seepage slope, 2003) rainfall events occurred, many producing high rainfall amounts (>1.52 cm) and intensities (>0.76 cm h^–1) that easily could have generated incidental transfers of applied P. Although the alternate hypothesis of dilution could help to explain runoff events that occurred during the lag period, 53% (seepage slope) to 71% (transportational midslope) of these events still had TP concentrations in surface runoff that were more than 2.5 times greater than the TP concentration we defined as chronically desorbed from soils (1.1 mg P L^–1). This suggests again that additional factors were at play beyond the simple dilution effects reported by Kleinman et al. (2006).

Runoff Generation Mechanisms and Hydrologic Connectivity
Differences in runoff generation mechanisms had important effects on P concentrations and loads at each landscape position. Infiltration excess runoff generation accounted for at least 89% of the runoff events from the seepage slope and transportational midslope (Table 4). These events typically generated small volumes of runoff with relatively high concentrations of P (Table 4) due to their proximity to the P sources of fields 1 and 2 (Fig. 1; Table 1). Infiltration excess surface runoff was also important at the south footslope, where it accounted for about 92% of the P loads (Table 4). Despite the importance of infiltration excess runoff to these sites, across all sites infiltration excess runoff accounted for only 37% (57% when the effects of Ivan are factored out) of the total P load observed during the study period as surface runoff from the north footslope outweighed those observed elsewhere.

At the north footslope, infiltration excess and saturation excess surface runoff events occurred with similar frequency (Table 4), and P loads in saturation excess runoff accounted for 88% of the total P load (Table 4). Saturation excess runoff from the north footslope was characterized by larger volumes but lower P concentrations than infiltration excess runoff (Table 4). The low concentrations observed from the footslope saturation excess events run contrary to the findings of Zheng et al. (2004) and Sánchez and Boll (2005), who observed that P concentrations in saturation excess runoff are higher than in infiltration excess runoff because saturation excess runoff water remains in contact with P-saturated upper soil layers for longer periods of time than does infiltration excess runoff water. In the current study, the large volumes associated with saturation excess runoff likely diluted P concentrations relative to those measured in infiltration excess runoff (Table 4).

Variability in P transport across the landscape was also expressed in the connectivity of surface runoff between landscape positions, which ultimately determines whether runoff from a field will directly affect downstream water quality (Sharpley et al., 2008). Over the 30-mo period of observation, 14 storms (15% of the total number of events) were identified where surface runoff was generated at the seepage slope or the transportational midslope but not at the north footslope, indicating a disconnect between upper and lower landscape positions. These storms accounted for a total of 0.06 g (27%) of the P load and point to a segmented conveyance mechanism for P in which it is transferred downslope but not to the stream. The fate of that P is unknown and requires further investigation because the absence of surface runoff from the footslope position does not mean that subsurface P transfers did not occur. However, these results do support the hypothesis that a substantial amount of the P transfer within this landscape occurs in a two-step conveyance process, with incidental transfers of manure and fertilizer P accounting for P translocation to downslope areas, which, over time, have the ability to remobilize that P to the stream via chronic mechanisms of soil P desorption. Therefore, even if incidental transfers of P in runoff from fields 1 and 2 do not entirely connect to the stream, the potential for this P to be remobilized from the footslope supports a management strategy that minimizes incidental transfers of P from upslope regions.

Implications
Results of this study affirm the critical source area concept of P management and offer important insight into the P-Index. The Pennsylvania P-Index (Weld et al., 2007) rates fields 2 and 3 as having medium to high risk of P loss from 2002 to 2004, reflecting a strong weight by the P-Index on source factors affecting incidental transfers of P. Indeed, incidental transfers appeared to be of relative importance to the P loads observed from sites associated with these fields, but the actual size of these loads was quite low. In contrast, field 4, despite having high transport factors, consistently had low P-Index scores over the study period due to the fact that neither manure nor fertilizer were applied there during the study. Landscape trends in runoff P loads observed in this study would support a relative P-Index rating for field 4 that was greater than those for fields 2 and 3. Results of this study clearly point to the potential to adjust the P-Index to better represent the importance of chronic diffuse transfers, such as by adjusting the representation of transport potential within the P Index or the relative weighting of soil P source factor in comparison with applied P source factors.

	Conclusions

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Phosphorus in surface runoff is affected by a variety of source and transport factors. Understanding the interaction among these factors is the key to managing water quality in agricultural watersheds. By monitoring natural rainfall-runoff events at different landscape positions with varying field management, this study sheds light on the interaction of runoff hydrology, field management, and landscape position on sources and transport of P at the hillslope scale.

Surface runoff varied by landscape position, with saturation excess runoff accounting for 19 times the volume of infiltration excess runoff from the north footslope but infiltration excess runoff dominating at the other landscape positions. Runoff differed significantly between south and north footslopes, coinciding with the extent of upslope soil underlain by a fragipan. Phosphorus in runoff was predominantly in dissolved reactive form (70%), with the highest concentrations associated with the upper landscape positions that were closest to fields serving as major sources of P. The highest concentrations of P in runoff were in summer and fall, coincident with applied fertilizer and manure and many of the low flows observed during the study period. However, the largest loads of P were from the north footslope, where runoff volumes were 24 times larger than from all other sites combined. Most of the P loads in runoff from the north footslope appeared to be related to chronic transfers of desorbed soil P explained by existing relationships between soil test P and DRP in surface runoff. Although runoff from the footslope likely contributed directly to stream flow and hence to stream water quality, 27% of runoff P from the upslope sites did not appear to connect with stream flow.

	ACKNOWLEDGMENTS

 
The authors are grateful to the staff of the USDA-ARS Pasture Systems and Watershed Management Research Unit for their contributions to this study. In particular, Mike Reiner, Todd Strohecker, Jim Richards, David Otto, and Terry Troutman were responsible for the success of the field study, including equipment construction and installation, site maintenance, and data collection. Laboratory analyses were ably conducted by Joan Weaver, MaryKay Lupton, Paul Spock, and Charles Montgomery. Lou Saporito oversaw database development.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 

• Aron, G., D.J. Wall, E.L. White, and C.N. Dunn. 1987. Regional rainfall intensity-duration curves for Pennsylvania. J. Am. Water Resour. Assoc. 23:479–485.[CrossRef]
• Bremner, J.M. 1996. Nitrogen-Total. p. 1085–1121. In D.L. Sparks (ed.) Methods of soil analysis. Part 3. Chemical methods. SSSA Book Ser. 5. SSSA and ASA, Madison, WI.
• Buda, A.R., P.J.A. Kleinman, M.S. Srinivasan, R.B. Bryant, and G.W. Feyereisen. 2009. Factors affecting surface runoff generation from an agricultural hillslope in central Pennsylvania. Hydrol. Processes 23:1295–1312.[CrossRef]
• Carpenter, S.R., N.F. Caraco, D.L. Correll, R.W. Howarth, A.N. Sharpley, and V.H. Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecol. Appl. 8:559–568.[CrossRef]
• Conacher, A.J., and J.B. Dalrymple. 1977. The nine-unit landsurface model: An approach to pedogeomorphic research. Geoderma 18:1–154.[CrossRef][Web of Science]
• Dougherty, W.J., N.K. Fleming, J.W. Cox, and D.J. Chittleborough. 2004. Phosphorus transfer in surface runoff from intensive pasture systems at various scales: A review. J. Environ. Qual. 33:1973–1988.[Abstract/Free Full Text]
• Dunne, T., and R.D. Black. 1970. An experimental investigation of runoff production in permeable soils. Water Resour. Res. 6:478–490.[CrossRef]
• Ghebreiyessus, Y.R., and J.M. Gregory. 1987. Crop canopy functions for soil erosion prediction. Trans. ASAE 30:676–682.[Web of Science]
• Gburek, W.J., and J.G. Broyan. 1974. A natural non-point phosphate input to small streams. p. 39–50. In R.C. Loehr (ed.) Proceedings of Agricultural Waste Management Conference. Rochester, NY. 25–27 March 1974. Cornell University, Ithaca, NY.
• Gburek, W.J., A.N. Sharpley, and H.B. Pionke. 1996. Identification of critical source areas for phosphorus export from agricultural catchments. p. 262–282. In M.G. Anderson and S.M. Brooks (ed.) Advances in hillslope processes, vol. 1. John Wiley & Sons, Chichester, UK.
• Gburek, W.J., B.A. Needelman, and M.S. Srinivasan. 2006. Fragipan controls on runoff generation: Hydropedological implications at landscape and watershed scales. Geoderma 131:330–344.[CrossRef][Web of Science]
• Haygarth, P.M., and S.C. Jarvis. 1997. Soil derived phosphorus in surface runoff from grazed grassland lysimeters. Water Resour. Res. 31:140–148.[CrossRef]
• Haygarth, P.M., and S.C. Jarvis. 1999. Transfer of phosphorus from agricultural soils. Adv. Agron. 66:195–249.[CrossRef]
• Haygarth, P.M., A.L. Heathwaite, S.C. Jarvis, and T.R. Harrod. 2000. Hydrological factors for phosphorus transfer from agricultural soils. Adv. Agron. 69:153–178.
• Heathwaite, A.L., and R.M. Dils. 2000. Characterising phosphorus loss in surface and subsurface hydrological pathways. Sci. Total Environ. 251/252:523–538.
• Howarth, R.W., A.N. Sharpley, and D. Walker. 2004. Sources of nutrient pollution to coastal waters: Implications for achieving coastal water quality goals. Estuaries 25:656–676.[CrossRef]
• Horton, R.E. 1933. The role of infiltration in the hydrologic cycle. Trans. Am. Geophys. Union 14:446–460.
• Kleinman, P.J.A., M.S. Srinivasan, C.J. Dell, J.P. Schmidt, A.N. Sharpley, and R.B. Bryant. 2006. Role of rainfall intensity and hydrology in nutrient transport via surface runoff. J. Environ. Qual. 35:1248–1259.[Abstract/Free Full Text]
• McDowell, R.W., and A.N. Sharpley. 2002. Phosphorus transport in overland flow in response to position of manure application. J. Environ. Qual. 31:217–227.[Abstract/Free Full Text]
• McDowell, R.W., and W. Catto. 2005. Alternative fertilisers and management to decrease incidental phosphorus loss. Environ. Chem. Lett. 2:169–174.[CrossRef]
• Mehlich, A. 1984. Mehlich-3 soil test extractant: A modification of Mehlich-2 extractant. Commun. Soil Sci. Plant Anal. 15:1409–1416.[CrossRef][Web of Science]
• Misselbrook, T.H., B.F. Pain, A.C. Stone, and D. Scholefield. 1995. Nutrient run-off following application of livestock wastes to grassland. Environ. Pollut. 88:51–56.[CrossRef][Medline]
• Murphy, J., and J.P. Riley. 1962. A modified single solution method for determination of phosphate in natural waters. Anal. Chim. Acta. 27:31–36.[CrossRef]
• Nash, D., M. Hannah, D. Halliwell, and C. Murdoch. 2000. Factors affecting phosphorus export from a pasture based grazing system. J. Environ. Qual. 29:1160–1166.[Abstract/Free Full Text]
• Needelman, B.A. 2002. Surface runoff hydrology and phosphorus transport along two agricultural hillslopes with contrasting soils. Ph.D. thesis. Pennsylvania State University, University Park, PA.
• Needelman, B.A., W.J. Gburek, G.W. Peterson, A.N. Sharpley, and P.J.A. Kleinman. 2004. Surface runoff along two agricultural hillslopes with contrasting soils. Soil Sci. Soc. Am. J. 68:914–923.[Abstract/Free Full Text]
• 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 an agricultural hill-land watershed. J. Hydrol. 220:626–673.
• Pionke, H.B., W.J. Gburek, and A.N. Sharpley. 2000. Critical source area controls on water quality in an agricultural watershed located in the Chesapeake Basin. Ecol. Eng. 14:325–335.[CrossRef][Web of Science]
• Preedy, N., K. McTiernan, R. Matthews, L. Heathwaite, and P. Haygarth. 2001. Rapid incidental phosphorus transfers from grassland. J. Environ. Qual. 30:2105–2112.[Abstract/Free Full Text]
• SAS Institute. 2004. SAS 9.1.3 Help and documentation. SAS Institute, Cary, NC.
• Sánchez, M., and J. Boll. 2005. The effect of flow path and mixing layer on phosphorus release: Physical mechanisms and temperature effects. J. Environ. Qual. 34:1600–1609.[Abstract/Free Full Text]
• Scholefield, D., and A.C. Stone. 1995. Nutrient losses in runoff water following application of different fertilisers to grassland cut for silage. Agric. Ecosyst. Environ. 55:181–191.[CrossRef]
• Sharpley, A.N. 1981. The contribution of phosphorus leached from crop canopy to losses in surface runoff. J. Environ. Qual. 10:160–165.[Abstract/Free Full Text]
• Sharpley, A.N., S.C. Chapra, R. Wedepohl, J.T. Sims, T.C. Daniel, and K.R. Reddy. 1994. Managing agricultural phosphorus for protection of surface waters: Issues and options. J. Environ. Qual. 23:437–451.[Abstract/Free Full Text]
• 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.
• Sharpley, A.N., J.L. Weld, D.B. Beegle, P.J.A. Kleinman, W.J. Gburek, P.A. Moore, Jr., and G. Mullins. 2003. Development of phosphorus indices for nutrient management planning strategies in the United States. J. Soil Water Conserv. 58:137–152.
• Sharpley, A.N., P.J.A. Kleinman, A.L. Heathwaite, W.J. Gburek, G.J. Folmar, and J.P. Schmidt. 2008. Phosphorus loss from an agricultural watershed as a function of storm size. J. Environ. Qual. 37:362–368.[Abstract/Free Full Text]
• Smith, D.R., P.R. Owens, A.B. Leytem, and E.A. Warnemunde. 2007. Nutrient losses from manure and fertilizer applications as impacted by time to first runoff event. Environ. Pollut. 147:131–137.[CrossRef][Medline]
• Srinivasan, M.S., M.A. Wittman, J.M. Hamlett, and W.J. Gburek. 2000. Surface and subsurface sensors to record variable source runoff generation areas. Trans. ASAE 43:651–660.[Web of Science]
• Srinivasan, M.S., W.J. Gburek, and J.M. Hamlett. 2002. Dynamics of stormflow generation: A hillslope-scale field study in east-central Pennsylvania, USA. Hydrol. Processes 16:649–665.[CrossRef]
• USEPA. 2007. National Water Quality Inventory: Report to Congress, 2002 Reporting Cycle. Office of Water, Washington, DC. EPA 841-R-07-001.
• Vadas, P.A., P.J.A. Kleinman, A.N. Sharpley, and B.L. Turner. 2005. Relating soil phosphorus to dissolved phosphorus in runoff: A single extraction coefficient. J. Environ. Qual. 34:572–580.[Abstract/Free Full Text]
• Wang, Y., D.R. Edwards, T.C. Daniel, and H.D. Scott. 1996. Simulation of runoff transport of animal manure constituents. Trans. ASAE 39:1367–1378.[Web of Science]
• Ward, R.C. 1984. On the response to precipitation of headwater streams in humid areas. J. Hydrol. 74:171–189.[CrossRef]
• Weld, J., D. Beegle, R. Bryant, W. Gburek, P. Kleinman, and A. Sharpley. 2007. The Pennsylvania phosphorus index: Version 2. Publications distribution center, Pennsylvania State University, University Park, PA.
• Withers, P.J.A., S.D. Clay, and V.C. Breeze. 2001. Phosphorus transfer in runoff following application of fertilizer, manure, and sewage sludge. J. Environ. Qual. 30:180–188.[Abstract/Free Full Text]
• Zheng, F.L., C.H. Huang, and L.D. Norton. 2004. Effects of near-surface hydraulic gradients on nitrate and phosphorus losses in surface runoff. J. Environ. Qual. 33:2174–2182.[Abstract/Free Full Text]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Buda, A. R. Articles by Feyereisen, G. W. PubMed PubMed Citation Articles by Buda, A. R. Articles by Feyereisen, G. W. Agricola Articles by Buda, A. R. Articles by Feyereisen, G. W. Related Collections Water Quality Watershed and Landscape Processes Phosphorus Nutrient Management Water Pollution