USGS - science for a changing world

Scientific Investigations Report 2009–5030

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2009–5030

Back to Table of Contents

Effects of Agricultural Practices

Hydrology

Knowledge of how water moves through a catchment is fundamental to understanding the effects of agricultural practices. One of the primary objectives of this work, therefore, was to develop a conceptual model of the sources, distribution, and movement of water within the DR2 catchment. A simple water balance (fig. 3), estimated from 2003 data, provides insight into the overall hydrology of the catchment and the processes that influence agricultural effects on the environment in this study area.

Irrigation water, imported from the Yakima River, is the principal source of water to the catchment, accounting for nearly 60 percent of the total inflow. This irrigation water is delivered through the Sunnyside Canal (fig. 1) and is transported throughout the catchment by a series of secondary canals, from which it is applied to crops and pasture. Rill irrigation—the predominant application method used in the catchment—typically results in 15 to 25 percent of the applied water running out the end of the furrows (U.S. Environmental Protection Agency, 2003; Eisenhauer and others, 2006; Schwankl and others, 2007). This tailwater is routed into surface or subsurface drains that eventually discharge to regional surface-water drains, such as DR2. Sprinkler irrigation—the other major water application method in DR2—produces no tailwater in most circumstances. Most canal water diverted into the irrigation water delivery system is applied to the land surface; however, a small fraction of the diverted water goes unused as part of the normal operation of the gravity-driven irrigation-water delivery network. This end-of-system “operational spill” also is routed into the regional drainage network. Additional unused water may enter the drainage system due to a lag time between requests for changes in the delivery amount and the implementation of the request, the onset of unexpected cool or wet weather, system malfunctions (such as pipe breaks), and various on-farm exigencies.

Other primary sources of water to the catchment are ground water that flows into the area from north of the Sunnyside Canal, precipitation, and leakage from the Sunnyside Canal. Together, these four major sources contribute approximately 180 cm of water per year to the catchment. Evapotranspiration and stream discharge account for nearly 90 percent of the total catchment outflow. The third important component of outflow is ground water.

Irrigation is a key component of the hydrology of the DR2 catchment (fig. 3). In fact, there is no evidence that a surface-water system existed in the catchment before irrigated agriculture began in the area more than a century ago (Payne and others, 2007). In addition to creating an extensive surface-water drainage network during the growing season, long-term irrigation has caused ground-water levels to rise, and the higher water table provides base flow that sustains the surface-water flow year-round in the southern part of the catchment.

A hydrograph of the discharge of DR2 near the catchment outflow (site W13) shows the influence of the annual irrigation cycle (fig. 4). Ground-water levels measured in wells throughout the catchment were consistently higher than the stage of the water in DR2, indicating that ground water flowed toward DR2 at all times. During the nonirrigation season (mid-October through mid-March), base flow in the stream is sustained by ground water, most of which originated as infiltrating irrigation within and upgradient of the DR2 basin. During the irrigation season (mid-March through mid-October), discharge at site W13 increased two- to threefold. The shape of the annual discharge hydrograph is defined by the controlled application of irrigation water, and because of the absence of major precipitation events, short-term variation in stream flow is small and occurs predominantly during the irrigation season, driven by crop demands and agricultural practices.

A conceptual model of subsurface flow in the DR2 catchment was developed on the basis of several lines of evidence, including ground-water levels, ground-water age estimates, the isotopic composition of water from selected sites, visual observations of flow in the surface drainage network, and results of numerical ground-water modeling and flow-path simulations (Leon J. Kauffman, U.S. Geological Survey, written commun., 2006). In this conceptualization, regional ground-water generally flows from north to south, but the local shallow system is more complex, consisting of small-scale subsystems driven by individual irrigation events and constrained by local topography and the extensive system of artificial drains (both surface and buried) that intercept shallow ground water and route it toward the surface-water drainage network (fig. 5). While the drain system promotes lateral flow to the stream, upward flux of deeper ground water through the streambed is limited, especially in the southern part of the catchment, by low-permeability streambed material. At the streambed transects, permeabilities ranging from 10-5 to 10-6 m/s were estimated by temperature flux modeling (Essaid and others, 2008). Finally, a considerable volume of deep ground water flows out of the catchment as ground water (fig. 3), not discharging to the surface system until it reaches Granger Drain or the Yakima River (fig. 1).

Water-level measurements in wells in the northern part of the catchment—at sites W24 and W25 (fig. 1)—consistently showed a downward gradient (fig. 6B-C). Several lines of evidence indicate that a downward vertical gradient also exists in the ground-water system near well W30. Although a second well at a different depth was not available to confirm it, the seasonal fluctuations in the ground-water level and temperature at W30 were similar to those in W24 and W25. In addition, age tracers (SF6 and CFCs) in water from W30 were consistent with concentrations in the atmosphere in the early 1980s, indicating little mixing of this water with either older or younger water. Finally, there are no surface water discharge zones in the area. All these factors are consistent with predominately downward vertical movement of ground water.

Ground-water levels in the northern part of the catchment fluctuated on an annual cycle (fig. 7A-C), with the lowest levels occurring in late spring to early summer, and the highest levels occurring in late summer to early fall in a delayed response to irrigation. The annual range in water levels in these wells was 1–2 m. This range is larger than was observed downgradient because the relatively thick unsaturated zone in the upper catchment prevents discharge to surface-water outlets. Although the water table rises and falls in the upper portion of the catchment in response to infiltrating irrigation water, the response to individual irrigation events is dampened during the movement of the water through the thick unsaturated zone. Water temperatures in these wells also fluctuated on an annual cycle, but the seasonal variation was less than 1°C (fig. 7A-C), again due to the insulating and moderating influence of a relatively thick unsaturated zone.

In the southern part of the catchment, at sites W20 and W21 (fig. 1), as well as at neighboring site W32, ground-water levels showed an upward gradient (fig. 6F-H). At these wells, water levels sometimes fluctuated on a daily basis, but showed no seasonal pattern (fig. 7D-E). The daily fluctuations were likely a response to individual irrigation events and (or) evapotranspiration, evident in these wells because the water table in this part of the catchment is relatively shallow. The shallow water table also was reflected in the water temperatures in these wells. Relative to those in the upper catchment, temperatures varied considerably, with summer maximum temperatures approximately 5°C greater than winter minimum temperatures (fig. 7).

Site W23 is in the midcatchment, where the predominant direction of vertical flow in the ground-water system transitions from downward to upward. Comparing levels measured in sites W23p and W23q (fig. 8) shows that the vertical gradient is generally upward, but periodically reverses to downward during the irrigation season. Data from this well demonstrate the response of the ground-water system to individual irrigation events and illustrate changes in the flow system in response to a change in irrigation method. Rill irrigation was used near these wells during 2003. In 2004, the farmland immediately upgradient of these wells was converted to sprinkler irrigation. During 2003, the frequency of head reversals was greater and the magnitudes of downward gradients were greater. Water levels in these wells also illustrate the general seasonal effect of irrigation—the vertical gradient diminished when canal deliveries ceased in mid-October 2003, and approached zero by the end of the end of the winter. Finally, corresponding increases in water temperature (fig. 8) indicate that gradient reversals in the ground water at site W23 are not due simply to a pressure response, but rather to warmer water from the surface or near surface actually being transported into the saturated zone and raising the local water table.

Estimated dates of ground-water recharge, based on interpretation of the CFC, SF6, and 3H data (Green and others, 2008), generally are consistent with the conceptual model of subsurface flow developed from ground-water levels (fig.  9). Data from wells in the midcatchment (sites W23 and W24) suggest that this is a mixing zone of waters recharged at different times. Age tracers from site W24q are consistent with this water being a mixture of water recharged around 2000 and water recharged around 1970. Age-tracer data from site W23q are consistent with water in this well being predominantly water recharged around 1980 but indicate contributions from water recharged prior to the mid-1960s. These mixtures are consistent with vertical mixing that would result from periodic vertical gradient reversals in the midcatchment and also with the conceptual model of a dynamic shallow flow system highly influenced by irrigation applications and drainage modification sitting atop a reservoir of considerably older, slower moving water. Throughout the catchment, water from the shallow water-table wells contained at least a component of recently recharged water, reflecting the movement of excess irrigation water into the underlying saturated zone.

The effects of artificially enhanced drainage in the southern part of this system can be seen by comparing the hydrographs of wells at sites W20p and W21p (fig. 7D-E). Short-term fluctuations in the water level at W20p show the effects of individual irrigation events in the adjacent pasture. By contrast, the water level at W21p shows little variation, likely due to the damping effect of the shallow drain that surrounds this field and efficiently shunts excess irrigation to the surface-water system. Age-tracer data from W21p (fig.  9), which indicate that the shallow ground water here was not recently recharged, support this reasoning.

The stable isotopes of water from selected sites (fig. 10) also corroborate the conceptual model of flow in DR2. From late April to late August, the canal water becomes enriched in the heavier isotopes, oxygen-18 (18O) and deuterium (2H). Samples from W23p and W24p are isotopically indistinguishable from late season canal water, consistent with this shallow ground water being recently applied irrigation. W21s is somewhat more depleted in the heavy isotopes than the other wells sampled. This was the deepest and oldest ground water sampled (fig. 9), and this isotopic signature likely reflects a contribution from older, regional ground water. W25p is slightly enriched in the heavier isotopes relative to other samples; these isotope values suggest a source of water similar to other nearby ground water (such as that from W23p, W24p, and W25q) that has undergone additional evaporative losses.

The isotopic composition of streambed water from W63c differs considerably from all other water sampled. That sample is much more depleted in 2H and 18O than other water samples collected for this study, indicating that the shallow subsurface water near the catchment outlet is derived from a different source than other water sampled. Particle-tracking simulations suggest that the deepest, longest ground-water flow paths in the catchment discharge near W63c and that water from the streambed there includes a significant component of the oldest ground water in the catchment. Tritium data (not shown) for the three streambed sites sampled (W51c, W55b, and W64c) also indicate that water from these sites includes a substantial component of relatively old water, recharged more than 50 years ago.

The isotopic signature of water from the DR2 drain (W13) is similar to that of shallow ground water and water from the Sunnyside Canal, and the shift in isotopic signature from spring to summer in water from the drain mimics that seen in water from the canal. During both seasons, however, the water from DR2 drain was slightly depleted in the heavier isotopes relative to corresponding canal water samples, which is consistent with a small contribution from deep ground water (fig. 10).

Water Chemistry

Water chemistry data were collected at sites representing several environmental compartments (fig. 1). Analysis of samples from the Sunnyside Canal (W15) indicates the composition of imported irrigation water prior to distribution in the catchment. Samples from a field drain (W42) provide an example of the chemistry of excess irrigation runoff from a single field—in this case likely a mixture of overland flow and shallow soil drainage. Samples from wells (W20–W32) reveal the characteristics of ground water at various locations throughout the catchment, both near the water table and somewhat deeper, but all in the shallow part of the flow system. Sites in two streambed transects in the lower reach of DR2 (W51–W65) show the composition of ground water just prior to its discharge into the stream. Finally, samples from near the DR2 catchment outlet (W13) reveal the composition of surface-water discharge from the catchment.

As discussed earlier, virtually all of the surface-water discharge from the catchment can be attributed to excess irrigation. As this water passes through the catchment, its chemistry is altered by biochemical and geochemical reactions at the land surface, in the soil, and in the ground-water system.

Fertilizers and other soil amendments used in the DR2 catchment are important sources of N and P, as well as potassium, calcium, magnesium, sodium, chloride, and sulfur. Cattle manure is applied as a fertilizer and soil amendment to many crops in the area either as liquid slurry or as solids, and, based on the classification scheme of Kendall (1998), 15N and 18O isotopes of nitrate (fig. 11) suggest that manure may be a major source of the nitrate measured in waters throughout the catchment.

Mineralogical analyses of subsurface material (sampled during well drilling) show its composition is predominantly quartz, plagioclase, potassium feldspar, and calcite. Through weathering processes, these minerals are abundant sources of silica, calcium, sodium, and potassium. Dolomite and hornblende, which contain magnesium, also are common in the area. Finally, smectite, which has a high cation-exchange capacity, was an important component of the colloidal fraction of most samples analyzed, and likely plays a role in water chemistry.

Comparing the chemistry of water from a number of environmental compartments and locations provides insight into the processes that affect the quality of water discharging from this catchment:

Irrigation water.— from the Sunnyside Canal had the lowest concentrations of major solutes and dissolved nutrients of any compartment sampled (figs. 1213). Because the canal is the major source of water to the catchment, the chemistry of water from other environmental compartments gives an indication of how this canal water is altered as it passes through the catchment.

Field drain.—The field drain flowed year-round, and both the discharge and chemical character of the water fluctuated with time. During winter, the major solute chemistry of water from the drain was similar to that of water from nearby well W21p, suggesting that winter flow in the drain was discharging ground water. Over the course of the irrigation season, the chemistry of the water from the field drain varied considerably (for example, N ranges from less than 1 to more than 23 mg/L; silica ranges from 13 to 47 mg/L), suggesting changing contributions from several sources, including ground water, shallow soil water, and irrigation tailwater. This composite discharge from the drain consistently had higher concentrations of major solutes and dissolved nutrients than water from the canal (figs. 1213). Concentrations of the major solutes calcium, magnesium, potassium, sodium, chloride, sulfate, and silica, were inversely related to drain flow rates (R2 = 0.4–0.6). Concentrations of nitrate also tended to be lower at higher flow rates (R2 = 0.3), but although the highest concentrations of soluble reactive P (SRP) were measured during low-flow conditions, there was no apparent relationship between SRP and flow (R2 < 0.1). The major solutes, as well as nitrate, also were inversely correlated with concentrations of suspended sediment in water from the field drain, though relationships were weaker than for discharge (R2 = 0.2–0.3). The highest concentrations of SRP also occurred when concentrations of suspended sediment were low, but again, the relationship between SRP and suspended sediment was not significant (R2 < 0.1). Data from Ebbert and others (2003) and Fuhrer and others (2004) point to a relatively constant contribution of SRP from ground water in this area. The poor correlation between SRP and suspended sediment may reflect differing mixtures of shallow ground water, soil water, and overland flow, in which concentrations of suspended sediment and dissolved constituents can vary considerably, depending on the amount of time since the field was last tilled and the timing of sample collection relative to individual irrigation events. Longer contact times between soil and water at lower flow rates may also be a contributing factor to the patterns observed. Durand and others (1999) noted that increased depth and velocity of overland flow (such as would occur during times of high drain discharge) tended to decrease concentrations of solutes.

Ground water.—The chemistry of the water from most individual wells varied little over the course of the study, but differences in water chemistry among wells were substantial. Relative to water from the canal and field drain, ground water typically had higher concentrations of major solutes. Although concentrations varied considerably among wells, water from the shallow wells typically had higher concentrations than that from deeper wells, suggesting that water-rock interactions in the subsurface may be a less important source of the major solutes than near-surface sources such as cattle manure, other fertilizers and soil amendments, and weathered soil.

The highest concentrations of N in ground water, which was present predominantly as nitrate, generally were detected in water from wells in the upper catchment (sites W24, W25, and W30). Concentrations of N in water samples collected from well W25p consistently exceeded 70 mg/L. This is probably the result of a complex interaction of factors, including the local ground-water flow system and local agricultural practices. Site W25 is at a local topographic high, and subsurface flow is likely to be directed downward rather than laterally. The very recent recharge date calculated for W25p (fig. 9) is consistent with relatively rapid, direct downward flow near the water table. Furthermore, this well is in an area where cattle manure, a potential source of considerable N, was applied liberally.

Concentrations of excess dissolved N2 in water from most wells was low (table 1). Dissolved N2 in excess of the concentration achieved through equilibration with the atmosphere plus excess air (Dunkle and others, 1993) indicates that nitrate previously present in the water has undergone denitrification—the process by which nitrate is transformed into N2. Thus, low concentrations of excess dissolved N2 in water from most wells indicate that denitrification is not a major process in this ground-water system. Green and others (2008) present the details of excess dissolved N2 calculations and discuss the limited denitrification in this ground-water system.

Streambed transects.—Similar to the findings for ground water, the chemistry of water from individual sites within the streambed transects varied little over the course of the study. However, the range in the data from each transect (figs. 1, 1114, table 1) shows that water chemistry differs over very small distances (tens of centimeters) in this system. Some of these differences may result from spatial variations in the composition of the streambed, but the variability observed also suggests that flow paths from different source areas are converging to discharge into this lower reach of DR2, and provides insight into differences between the chemistry of relatively deep ground water that discharges upward into the stream and that of the shallow ground-water system that flows laterally into the stream.

Compared to the ground water sampled, water from both streambed transects tended to have lower specific conductance and alkalinity, as well as lower concentrations of calcium, magnesium, and sodium (fig. 12). In contrast, concentrations of potassium, chloride, silica, and SRP tended to be similar to those measured in ground water (figs. 1213). These conditions indicate that differences between water in the streambed and ground water were not likely due to simple geochemical processing or dilution (from unapplied irrigation water, for example), but suggests that a component of the streambed water comes from a source distinct from other ground water sampled. We hypothesize that this source is relatively old ground water discharging from the deep, long ground-water flow paths mentioned previously.

The concentrations of nitrate were lower in water from the streambed than in most ground water sampled (fig. 14). In water samples collected from the upstream transect, concentrations of excess N2 consistently exceeded 2.5 mg/L, except at sites W62b and W61b (table 1). In water samples collected from the downstream transect, concentrations of excess N2 ranged from less than 1 to more than 6 mg/L (at site W54a). These data indicate that denitrification occurs in portions of the subsurface, but that conditions vary over very small distances. Values of 18O and 15N isotopes from nitrate collected from the streambed also suggest that denitrification occurs in this zone (fig. 11). (Puckett and others [2008] provide further details on denitrification in the streambed.) Although several lines of evidence indicate that denitrification occurs in this zone, the effects on nitrate concentrations in surface water are limited by the small flux of water involved. Several factors combine to constrain water flux through the streambed. First, as mentioned earlier, the permeability of the streambed is low; second, irrigation flow during the growing season raises the water level in DR2, which reduces the gradient from ground water to the stream; and third, enhanced shallow drainage throughout the catchment shunts ground water laterally toward surface drains and effectively bypasses deeper flow paths.

Conditions at sites W61b and W62b, in the shallow, western part of the upstream transect indicate that lateral flow to the stream from a distinct water source occurs in this area. Concentrations of nitrate here were higher than in the rest of the transect, but concentrations of excess N2 were somewhat lower, suggesting a source of nitrate-rich water, but only limited progression of denitrification (due to either reduced capacity or shorter residence time). These two sites also had higher specific conductance, alkalinity, calcium, magnesium, potassium, sodium, and total N (mostly present as nitrate), and lower concentrations of chloride relative to other sites in the transect. Furthermore, water temperature near these sites was periodically colder than water deeper in the streambed (sites W61c and W62c) or water near the bottom of the overlying stream (fig. 15). These data indicate that colder water was flowing laterally toward the transect at middepth and provide evidence of lateral ground-water flow to the stream.

Catchment outlet.—The chemistry of the water at the catchment outlet is consistent with the water being a mixture of canal water, field drainage, ground water, and streambed discharge. Examining the chemistry at the catchment outlet in the context of the annual streamflow hydrograph provides further insight into processes in the catchment. Several of the major solutes (calcium, magnesium, sodium, chloride, sulfate, and silica) were inversely related to stream discharge. This pattern of lower concentrations during the irrigation season, when the highest discharges occurred in DR2, is illustrated by calcium and silica data (fig. 16A, B). Concentrations of nitrate in water from DR2 also were highest during the nonirrigation season, but concentrations during the irrigation season varied more than those of the major solutes (fig. 16C). In contrast, concentrations of SRP in the catchment outflow were highly variable, but generally increased during the irrigation season (fig. 16D).

Inverse End-Member Mixing Analysis

An inverse application of a simple end-member mixing analysis (EMMA) was used to help understand the differences in water chemistry between irrigation season and base-flow conditions in DR2 and to gain insight into the transport of agricultural chemicals in the catchment. As discussed in the previous sections, flow in DR2 was more than twice as high during the irrigation season than during base-flow conditions (in the winter), and the concentrations of most solutes were diluted during the higher-flow period. Of the major solutes and nutrients, only concentrations of SRP were higher during the irrigation season.

Conceptually, the water in DR2 during the irrigation season can be described as a mixture of ground-water base flow and a seasonal component, referred to here as “seasonal agricultural flow” (SAF). This conceptualization, modified from Domagalski and others (2008), can be expressed mathematically as

Figure - refer to figure caption for alternative text description (1)

Comparing and contrasting the chemical characteristics of the hypothetical SAF with those of canal water gives insight into the composite effects of agriculture in the DR2 catchment on water quality.

For a conservative solute, the mass balance can be expressed as

Figure - refer to figure caption for alternative text description (2)

where Ci-DR2 , Ci-bf , and CSAF are the median irrigation season concentrations of the solute in DR2, ground-water base flow, and SAF, respectively. Combining equations 1 and 2 and solving for CSAF yields

Figure - refer to figure caption for alternative text description (3)

To solve this equation for individual solutes, the median discharge and concentrations of solutes measured at site W13 between mid-June and mid-September 2004 were used to approximate Qi-DR2 and Ci-DR2 . Hydraulic head data collected in the stream and in nearby ground water (fig. 7) showed that the average hydraulic gradient from ground water toward DR2 during the irrigation season was only about 80 percent of the gradient measured during the nonirrigation season. On the basis of these data, Qi-bf  was approximated as 80 percent of the median discharge measured at site W13 during the nonirrigation season, from mid-October 2003 through mid-March 2004. Finally, making the assumption that the concentrations of solutes in base flow were similar during the irrigation and nonirrigation seasons, the medians of solute concentrations measured at site W13 between mid-October 2003 and mid-March 2004 were used to approximate Ci-bf .

The composition of SAF (table 2), calculated on the basis of equation 3, serves as a tool for understanding source and transport processes in the DR2 catchment. Conceptually, SAF is a hypothetical composite of excess irrigation water and includes system spill (water that is transported directly from the irrigation delivery system to the surface-drainage system, without being released to the landscape), tailwater, overland flow, and applied irrigation water that infiltrates and then travels by way of shallow, short subsurface flow paths (subsurface drains, for example) to the surface-water system without being incorporated into the ground-water system. SAF is, therefore, water from the canal delivery system that passes through the catchment into DR2 during a single growing season. Comparison of the chemical characteristics of this hypothetical water to those of canal water provides insight into the net composite effect of current agriculture on water quality in the DR2 catchment. Furthermore, because nearly all ground-water base flow in the catchment can be attributed to irrigation over the past century (Payne and others, 2007), differences between SAF and base flow are attributable to changes that result from passage through the subsurface and (or) changes in agricultural practices over time. Finally, a comparison of SAF characteristics to those of water from the field drain enables a further understanding of how closely the net effect of processes in the single field sampled approximates the net effect of processes in the catchment as a whole.

SAF compared to canal water and base flow.—Calculated concentrations of most solutes in the hypothetical SAF were intermediate between concentrations measured in water from the canal and those measured in water from DR2 during base-flow conditions (table 2). For several major solutes (calcium, magnesium, potassium, sodium, and silica), concentrations calculated for SAF were two to four times greater than concentrations in canal water, but only about 40–50 percent of the concentrations measured in base flow samples (fig. 17). These differences suggest that excess irrigation water picks up a considerable load of solutes as it travels through the catchment, even when following the relatively rapid surface and near-surface transport pathways postulated for SAF. However, subsurface flow paths that discharge as base flow are responsible for the greatest contribution of solutes to DR2. Estimates of recharge dates for the ground water sampled (fig.  9) indicate that time spent in the subsurface is on the order of several years to several decades. Therefore, in addition to chemical changes in the water that result from its passage through the subsurface, some of the differences between SAF and base flow may be the result of changes in agricultural practices over the past several decades.

In contrast to the major solutes, the calculated concentration of nitrate in SAF was nearly an order of magnitude greater than the median concentration measured in canal water (fig. 17A), but only about 30 percent of the median concentration measured in base flow. The SRP concentration in SAF was more than six times greater than the median measured in canal water (fig. 17D), and twice the median concentration measured in base flow.

SAF compared to field drain discharge.—SAF concentrations calculated for most of the major solutes, as well as nitrate, were similar to the median concentrations measured in water from the field drain (W42). This similarity indicates that with respect to most major solutes (calcium, magnesium, potassium, sodium, and silica), the net effect of sources and transport processes in the field from which drainage was sampled was similar to the net effect of sources and processes throughout the catchment as a whole. At both the field scale and the catchment scale, soil-water interaction and dissolution of applied fertilizers and soil amendments probably accounted for virtually all of these solutes.

In contrast to the major solutes mentioned above, concentrations of SRP, chloride, and sulfate were dissimilar in SAF and the field drain. Concentrations of SRP, chloride, and sulfate calculated for SAF were approximately two times the median concentrations measured in water from the field drain. These differences indicate that the sources of these solutes and (or) the processes affecting their transport and fate in the catchment as a whole are not closely approximated by sources and processes governing water quality at site W42.

Given the inherently complex, highly variable nature of field-scale runoff processes (Durand and others 1999; Langlois and Mehuys, 2003; Cerdan and others, 2004; Kurz and others, 2005), the variety of cropping, irrigation, tillage, and drainage practices used across DR2, and ongoing nearstream and instream biochemical processes, even the concentrations of SRP, chloride, and sulfate calculated for SAF are remarkably similar to the median concentrations measured in water from the field drain.

Back to Table of Contents

AccessibilityFOIAPrivacyPolicies and Notices

Take Pride in America logoUSA.gov logoU.S. Department of the Interior | U.S. Geological Survey
URL: http://pubs.usgs.gov/
Page Contact Information: Publications Team
Page Last Modified: Thursday, 10-Jan-2013 19:27:11 EST