JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grimaldi, C.
Right arrow Articles by Rouault, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grimaldi, C.
Right arrow Articles by Rouault, F.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Grimaldi, C.
Right arrow Articles by Rouault, F.
Related Collections
Right arrow Ground Water Quality
Right arrow Surface Water Quality
Right arrow Water Quality
Right arrow Watershed and Landscape Processes
Right arrow Nutrients
Published in J. Environ. Qual. 33:994-1001 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Surface Water Quality

Stream Nitrate Variations Explained by Ground Water Head Fluctuations in a Pyrite-Bearing Aquifer

C. Grimaldi*, V. Viaud, F. Massa, L. Carteaux, S. Derosch, A. Regeard, Y. Fauvel, N. Gilliet and F. Rouault

UMR INRA-ENSA, Sol-Agronomie-Spatialisation, 65 rue de Saint-Brieuc, CS 84215, 35042 Rennes Cedex, France

* Corresponding author (grimaldi{at}roazhon.inra.fr).

Received for publication December 12, 2002.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
In the context of agricultural nitrogen excesses in northwestern France, pyrite-bearing weathered schist aquifers represent important hydrological compartments due to their capacity to eliminate nitrate . Under oxygen-free conditions, nitrate is reduced simultaneously with the oxidation of pyrite leading to the release of sulfate . The aim of the present study is to identify the hydrological conditions under which the weathered schist ground water influences the stream water chemistry, leading to a decrease in NO3– concentration. We measured the ground water head on a small catchment over weathered schist, near the bank and under the streambed, and analyzed the chemical composition of the ground water as well as the stream water on both seasonal and storm-event timescales. Using SO42– as a tracer of the weathered schist ground water, we showed that ground water inflow caused a decrease of NO3– concentration in the stream during the autumn as well as during storm events in spring and summer. In summer, the NO3– concentration was controlled by the sources of the stream, and in winter by the shallow ground water inflow. The effect of the weathered schist ground water on the NO3– depletion remained relatively limited in time. This effect persisted into late autumn as long as the NO3––rich shallow ground water did not feed the stream. The duration and intensity of the effect would be extended by decreasing the shallow ground water inflow, which depends on climate as well as the presence of landscape features such as hedges and buffer zones.

Abbreviations: WB, bankside water-table observation well • WS1 and WS2, midstream water-table observation wells


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
MANY RIVERS of the Armorican Massif (northwestern France) and the ancient massifs of Western Europe drain catchment areas made up of Brioverian schist substrate with variable pyrite content (Kölle et al., 1985; Pauwels et al., 2000). These rivers are partly fed by ground water flowing through weathered schist aquifers. In the context of agricultural pollution of rivers, pyrite-bearing aquifers attract a certain interest owing to their capacity to remove nitrate. In the absence of oxygen, the nitrate is reduced coupled with the oxidation of sulfur in the pyrite (formula: FeS2) (Kölle et al., 1985; Robertson et al., 1996; Pauwels et al., 1998). This reaction is mediated by the bacterium Thiobacillus denitrificans:

[1]

The eutrophication of watercourses, lakes, and estuaries is a function of the nitrate load of streams. Nitrogen exported by the streams draining northwestern France amounts to 3700 kg km–2 yr–1, which is equivalent to an average nitrate content of 0.8 mmol L–1 (Aurousseau, 2001). Understanding the factors determining stream nitrate variations will lead to the application of water quality protection or restoration measures better adapted to the various types of functioning of catchment areas. The inflow from pyrite-bearing aquifers can lower the nitrate content of stream waters. Moreover, ground water studies contribute to our knowledge of the exchanges between surface and ground waters. Such exchanges can affect the biological functioning of the "hyporheic zone," an active and dynamic ecotone of the stream ecosystem (Pionke et al., 1988; Cirmo and McDonnell, 1997; Boulton et al., 1998).

The objective is to study the hydrological and geochemical dynamics of ground water in a pyrite-bearing aquifer, and its influence on lowering the nitrate content in a small stream located in the Armorican Massif. To this purpose, we measured the hydraulic head of the ground water next to and under the stream, and analyzed the chemical composition of the ground water as well as the stream water. We were interested in the seasonal and interannual variations of these parameters, as well as their episodic variations during storm events. By monitoring the dynamics of ground water head, coupled with the use of SO42– as a tracer, we attempt to show under which hydrological conditions the chemical composition of the stream can be influenced by the ground water.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Location of the Catchment Area
The study was performed in the catchment area of a second-order stream (Strahler, 1957) called Le Moulinet. This stream is a tributary of the Oir, itself a tributary of the Sélune, which flows into the Baie du Mont Saint-Michel (northwestern France) (48° N, 1° W) (Fig. 1) .



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 1. Sketched map showing location of Le Moulinet catchment and study site. The term D denotes the drain outlet, while S denotes the sources of the stream. (a) Ground water level survey and sampling site with wells under the stream bed (WS1 and WS2) and on the bank (WB). (b) Details of the ground water level device.

 
The catchment is situated in a region of moderately intensive agriculture, mostly dairy cattle farming. In 1995, 50% of the agricultural area was under pasture. Winter cereals and maize (Zea mays L.) are grown especially on the hillslopes. Ninety-three percent of the plots bordering stream banks are occupied by grassland grazed by cows a few months per year, and 6.5% by crop succession including maize (Baudry and Thenail, 2004). The fertilization is mainly inorganic, with 39% of the grassland receiving more than 120 kg N ha–1 yr–1.

The bedrock is made up of Brioverian schist, a geological substratum very commonly encountered in the Armorican Massif. Aeolian silty loess forms a covering of variable thickness on top of the weathered schist (Langevin et al., 1984). Because of the impermeability of the substratum, the ground water bodies are notably superficial, particularly in the valleys. As a result, waterlogged soils are found in the valleys while the hillslopes are occupied by well-drained soils. The substratum making up the bed of the studied stream is composed of weathered schist. This material is extremely compact and its gray-blue color indicates the presence of highly reducing conditions. The weathered schist in the streambed is covered by unconsolidated sediments of variable thickness (approximately 0.1 m). Near the stream, a colluvial–alluvial zone is restricted to a width of a few meters (Chaplot et al., 2000). In this riparian zone, the gray-blue weathered schist passes upward, at a depth of about 1 m, into loamy horizons that also show evidence of reducing conditions. This material becomes slightly more organic within 0.5 m of the ground surface.

The temperate maritime climate is characterized by a mean annual rainfall of 1058 mm (data from Météo-France records, 1977–2000, at about 10 km from the site), and precipitation of moderate intensity is distributed homogeneously throughout the year, being slightly more frequent in autumn and winter (September–March). Evapotranspiration leads to a hydric deficit that is sometimes well marked in spring and summer (April–August).

Study Site, Instrumentation, Sampling, and Analysis
The studied zone is located halfway along the course of the Moulinet stream, with a drained catchment surface area of 4.6 km2 (Fig. 1). The width of the stream here is about 1 m and the average depth in baseflow conditions is 0.1 to 0.2 m. Its gradient is very gentle, about 2 to 3%, and both banks are occupied by meadows with some poplars (Populus spp.) as riparian vegetation on the left side of the stream.

To measure the ground water head in the weathered schist, two water-table observation wells (WS1 and WS2) were installed in midstream 30 m away from each other, while a third was placed on the right bank (WB) in the bordering meadow 2 m away from the stream (Fig. 1). Each water-table well comprised a PVC tube 0.05 m in diameter, with a screened zone toward the base made up of slits over a depth interval of 0.8 m. The PVC tube was inserted into a hole of equivalent diameter, which was excavated with a mechanical corer due to the highly compact nature of the weathered schist. The screened zone of the tube was located in weathered schist at between 0.3 and 1.1 m depth beneath the streambed, and between 0.9 and 1.7 m beneath the soil surface on the bank to remain under the water-table level at all times. In fact, weathered schist is directly present beneath a few centimeters of sediments in the stream and under nearly 1 m of organic and alluvial deposits on the bank (Chaplot et al., 2000). The measurement array was installed in mid-February 2001. The water-table level was measured and recorded with an electronic water-level probe (Orphimèdes OTT, Aix en Provence, France). The probes were manually checked every 15 d. A slight drift (a few centimeters) was sometimes observed, requiring a readjustment of the recorders. The water-table level is expressed in comparison with the elevation of the streambed, at the location of each midstream well or at the least distance from the bankside well. The reference elevation is the same for the midstream well WS1 and for the bankside well WB.

To analyze the chemical composition of the weathered schist ground water, samples were collected from all three wells every 15 d between April 2001 and April 2002. The wells were protected from the rain by a stopper that was opened during each water sampling. Water was sampled by suction with a syringe. During each sampling (300 mL), the ground water level fell by about 15 cm. In the midstream wells, the water level returned to its initial height in less than 1 h up to about 12 h, whereas in the bankside well, the water level recovered much more slowly, 2 d on average (Fig. 2) . For this reason, we did not systematically follow the standard protocol that recommends evacuation of several well volumes before sampling. Nevertheless, on several occasions in winter when the well filling was more rapid, we sampled the ground water (i) without purging and (ii) following a complete purge of the well and after waiting for a filling time of one or a few hours. Nitrate remained below the detection limit in samples taken according to both protocols. The sulfate concentrations varied by less than 10%. In this study, we also present some analyses of the ground water flowing from the outlet of an agricultural drain network located near the site (Fig. 1) and sampled at various times toward the end of 1996. The drain network is composed of buried PVC pipes converging toward the same outlet. Since this period was particularly dry, we can assume that the drain collected weathered schist ground water rather than shallow ground water.



View larger version (50K):
[in this window]
[in a new window]
  		Fig. 2. (a) Daily rainfall. Variations of stream level and weathered schist ground water head in (b) the well under the stream bed (WS1) and (c) the well on the bank (WB).

 
To compare the chemical composition of stream water with the ground water, we collected samples of stream water on the site at the same dates as the well sampling (i.e., every 15 d between April 2001 and April 2002). During the monitoring from October 2001 to April 2002, daily stream water sampling was also performed by means of an automatic sampler.

To compare the stream water monitoring performed over one year with the interannual variability, we also present the results of a twice-monthly sampling campaign of stream water chemistry on the same site and in headstream sources at different seasons between December 1994 and September 2000. This period is characterized by two distinctly drier years in 1996 and 1997 (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Annual rainfall (data from Météo-France, Saint Hilaire du Harcouët).

 
We performed measurements of the electrical conductivity, pH, temperature, and dissolved oxygen content for all the water samples while in the field. The dissolved oxygen content was measured on an aliquot out of contact with the air and under constant agitation using a magnetic stirrer. The samples were placed in ice until being brought to the laboratory, then filtered at 0.45 µm and stored at 4°C before analysis. The concentrations of NO3–, Cl–, and SO42– were measured using ionic chromatography (Dionex, Sunnyvale, CA) with precisions of 3, 5, and 5%, respectively.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Weathered Schist Ground Water Head
Seasonal Variations
The ground water level measured in the wells midstream and on the bank is higher than the stream water surface during most of the year (Fig. 2). There is an upward gradient and the ground water is confined because of the very low hydraulic conductivity of the aquifer. The seasonal dynamics of the ground water head are well marked during the study period: the water-table level attains its maximum height in winter (March) and in early spring (April). It falls from May onwards and reaches its minimum at the end of August, becoming lower than the bed of the stream in each well. From September, the water table level increases once again. This dynamic behavior is appreciably the same in all three wells. In well WS2, the level of the ground water in winter is closer to the stream water surface, but also clearly falls in summer (data not presented).

Storm-Event and Daily Variations
In all seasons, the ground water head shows a clear increase in the midstream wells during rains, simultaneously with the rise in the level of the stream water surface (Fig. 2 and 3a) . These increases are of variable intensity in the different wells, while the effect is almost non-existent on the bank.



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 3. (a) Daily variations of stream levels (thick, solid line) and water-table levels (thin, solid line for under the stream bed [WS1 and WS2] and dashed line for on the bank [WB]) during winter high-flow events. (b) Daily variations of water-table level under the stream bed (WS1) during summer.

 
In summer, except during rainy periods, the ground water head displays daily variations of very strong amplitude (up to 0.3 m) in the midstream wells, but much less marked on the bank (Fig. 2). The minima are attained in the late afternoon and the maxima at the beginning of the morning (Fig. 3b). By performing manual measurements in the field during the course of one day, we checked that these results were due to variations of the water-table level itself rather than artifacts caused by variations in temperature affecting the apparatus. In other studies, diurnal fluctuations of ground water level or stream flow are commonly observed during dry periods and are attributed to the daily evapotranspiration of riparian vegetation (Kobayashi et al., 1990; Bren, 1997). Poplars occur as riparian vegetation at the present study site. These trees take up large amounts of water, with a mean consumption of 40 L d–1 individual–1 (Hinckley et al., 1994). Jaworski (1968) observed a ground water diurnal fluctuation of 0.17 m in sandy soil under a poplar nursery. Kobayashi et al. (1990) showed that stream water chemistry can also vary with the same periodicity during dry periods. In our study, water was always sampled at about the same time in the day (i.e., the .

Nitrate and Sulfate Contents in Weathered Schist Ground Water
Twice-monthly sampling of the weathered schist ground water in the three wells generally yields NO3– contents below the detection limit (5 µmol L–1), while stream water contents vary between 0.4 and 0.8 mmol L–1 during the same period (Fig. 4) . Well WS2 exhibits high NO3– concentration only in summer, with values close to the stream water content. The contents of SO42– in the ground water are generally higher than in the stream at the same time, except during the autumn. In summer, well WS2 also presents SO42– concentration close to the stream content. We believe that the piezometer WS2 did not maintain a good watertight seal around the PVC tube. A small defect in the seal quality does not cause a problem as long as the ground water hydraulic gradient is high and directed upward. In summer, however, the ground water head falls sharply beneath the streambed. At that time, stream water probably infiltrates around the PVC tube to reach the screened zone of the well.



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 4. Bimonthly variations of NO3– and SO42– concentrations in stream water (-{circ}-) and weathered schist ground water in the three wells (WB, -{diamondsuit}-; WS1, -{blacktriangleup}-; WS2, -{blacktriangledown}-).

 
The absence of nitrate and the high contents of sulfate in the ground water were confirmed by another ground water sampling campaign performed in 1996 (Grimaldi and Chaplot, 2000) and by sampling at the drain outlet. At this place the SO42– contents range up to 3.5 mmol L–1, accompanied by high contents of cations (between 1 and 2 mmol L–1 for Ca2+, Mg2+, and Na+) and a lowering of pH to 4.5 (Fig. 5) .



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 5. Increase of H4SiO4, Ca2+, Mg2+, K+, and Na+ concentrations and decrease of pH versus enrichment of SO42– in the weathered schist ground water at a drain outlet at different dates in late 1996.

 
The absence of nitrate and the high sulfate contents in the weathered schist ground water are attributed to oxidation of pyrite coupled with denitrification (i.e., Eq. [1]). The weathered schist is completely gleyified (Chaplot et al., 2000), which indicates the existence of reducing conditions required for such a process. In the drain samples, the increase in SO42– concentration is accompanied by strong acidity and high cation contents (Fig. 5). At the drain outlet, Fe2+ released by Eq. [1] is probably reoxidized involving Eq. [2] (Kölle et al., 1985; Pedersen et al., 1991):

[2]

Production of H+ then favors hydrolysis reactions that release silica and cations.

Nitrate and Sulfate Contents in Stream Water
The various stream water sampling campaigns show storm-event, seasonal, and interannual variations of the NO3– and SO42– contents, with occasional development of a negative correlation between these two anions (Fig. 4, 6, and 7) .



View larger version (46K):
[in this window]
[in a new window]
  		Fig. 6. Seasonal rainfall and potential evapotranspiration (PET) with variations of stream discharge, NO3–, and SO42– contents in stream water based on twice-monthly monitoring between 1995 and 2000.

 


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 7. Daily rainfall and Cl–, NO3–, and SO42– variations in stream water.

 
Seasonal and Interannual Variations
During the twice-monthly monitoring of the stream water chemistry between April 2001 and April 2002 (Fig. 4), the SO42– contents at first show rather little variation, with low levels around 0.07 mmol L–1. Contents then increase clearly in the autumn (September) and decrease again in winter (December). The NO3– contents follow an opposite trend.

The opposite evolution of the two anions as seen in autumn 2001 is observed each year at the same season during the monitoring performed between December 1994 and September 2000 (Fig. 6). In autumn, the SO42– contents are highest, particularly in autumn 1996, when they reach 1 mmol L–1. We also note some small transient increases in SO42– concentration in spring and in summer. Each increase in SO42– concentration is accompanied by a decrease in NO3– concentration (Fig. 6). For the rest of the time, SO42– contents are generally low and remain around 0.07 mmol L–1, while NO3– concentration varies according to the classical seasonal pattern developed in agricultural catchment areas of the Armorican Massif (Molénat et al., 2002) as well as in many regions of Europe and North America (Slack and Williams, 1985; Webb and Walling, 1985; Owens et al., 1991; Murdoch and Stoddard, 1992; Pionke et al., 1999). In these studies, the highest concentrations were measured in winter, and the lowest in autumn. The increase is rapid between autumn and winter, followed by a more gradual decline. Compared with this pattern, we observed different situations during the studied period: the NO3– contents measured during the more rainy winters of 1994–1995, 1998–1999, and 1999–2000 are higher than those measured during the three other drier winters. In the same way, the contents measured in the summers of 1999 and 2000, which were relatively rainy and followed rainy winters, are higher than levels during the first three drier summers.

Storm-Event Variations
Between September 2001 and March 2002, the NO3– contents in stream water drop sharply with each increase in flow caused by the rain (Fig. 7). These decreases in NO3– concentration are of the same order of magnitude in both autumn and winter. They are accompanied in autumn by increasing SO42– contents. On the other hand, SO42– falls to low values with little variation during the winter and the decreases in NO3– concentration at that time are associated with decreasing Cl– concentration.

Influence of Ground Water Inflow on Stream Water Sulfate and Nitrate Contents
The release of SO42– within the weathered schist aquifer allows us to use this anion as a tracer of the ground water. Indeed, SO42– concentration is significantly higher in the weathered schist ground water than the background level in shallow ground waters (Schnabel et al., 1993). The latter is related to the oxidation of organic matter or moderate inputs of fertilizer containing SO42–. In a neighboring stream with the same agriculture context but flowing over a granite substrate, the mean SO42– content of bimonthly stream water samples over 5 yr is 0.073 mmol L–1, with a standard error of 0.002 mmol L–1 (Grimaldi et al., 2000).

The influence of ground water on stream water chemistry, as reflected by the rise in SO42– concentration, is only visible under two conditions: (i) the ground water head must be relatively high to supply the stream flow, considering the very poor permeability of the weathered schist (Grimaldi and Chaplot, 2000) and (ii) the input from the ground water to the stream must be significant compared with the inputs from other compartments, in particular shallow ground water. These two conditions explain the seasonal and event-related variations of SO42– concentration in the stream water.

The SO42– contents in the stream increase primarily in autumn (Fig. 4, 6, and 7), when the ground water head increases along with the resumption of rainy weather and as long as the stream discharge remains low. From year to year, the influence of the weathered schist ground water on the stream water chemistry in autumn is more or less accentuated and sustained. The autumn increase of SO42– concentration in 1996 (Fig. 6) was thus exceptionally strong and sustained, due to moderate rainfall that limited the flow in the shallow ground water. The effect was more transient during the rainy periods at the end of 1998 and 1999. In winter, the stream water SO42– contents are low (Fig. 4, 6, and 7), although the ground water head is at a maximum (Fig. 2). During this season, shallow ground water largely dominates the inputs to the watercourse. In summer, the SO42– contents are also low because the water-table level of the weathered schist ground water drops below the streambed. During rain events, the rise in ground water level is reflected in a rapid and transient increase in stream water SO42– concentration when the discharge is rather low, that is, during storm events in spring and summer (Fig. 6) or in autumn (Fig. 7). The increase in SO42– contents is not at all visible during the winter storm events, when shallow ground water inflow contributes to the increase in discharge.

The increase of SO42– concentration in the stream water corresponds to events or seasons when there is significant contribution of the ground water to streamflow. Since the weathered schist ground water is both rich in sulfate and poor in nitrate, the drop in NO3– concentration in the autumn, as well as during storm events in spring and summer, is due to the inflow of ground water from the weathered schist. In winter, the ground water inflow is insufficient to reduce NO3– concentration, compared with the inflow from more concentrated shallow ground water. When NO3– concentration decreases during storm events in winter, this decrease is due to a dilution effect (Webb and Walling, 1985).

Effect of Sources on Stream Water Sulfate and Nitrate Contents in Summer
During the summer, when the water table drops beneath the streambed (Fig. 2), the weathered schist ground water does not seep into the stream. Discharge measurements performed along the entire length of the watercourse in October 1996 and April 1997 (Grimaldi and Chaplot, 2000) also indicate the absence of any ground water inflow into the stream along its downstream course. The question thus arises of the origin of the water feeding the watercourse in the absence of any ground water input. Indeed, although the summer discharge is low, particularly during the dry years 1996 and 1997, the stream never actually dries up. At such times, the stream is fed only by its sources (Fig. 1), which correspond to resurgences from a ground water body probably yielding a sustained flow. Moreover, these sources have in most cases been tapped as catchment works for the water supply of nearby hamlets. The depth of the water collecting system varies between 1 and 4 m, with a yield of between 40 and 125 m3 d–1 (Direction Départementale des Affaires Sociales et Sanitaires [DDASS] data, 1993–2001). Within this water catchment area, as recommended in the Armorican Massif (Marjolet et al., 2002), wooded vegetation zones covering about 1 ha protect the source catchment works against pollutants of agricultural origin.

The ground water feeding the headstream sources is less rich in sulfate than the weathered schist ground water sampled in this study, and has higher nitrate contents (Fig. 8) . Other data reported by the DDASS (from 1993–2001) mention nitrate variations between 0.08 and 1.2 mmol L–1 in the same sources. Ground water circulation takes place probably under relatively oxygenated conditions, in an aquifer different from the weathered schist studied farther downstream. The NO3– contents in the sources decrease appreciably between early spring (0.5–0.6 mmol L–1) and summer (0.2 mmol L–1) (Fig. 8). Nitrate consumption by the wooded vegetation in this headwater area may explain the low NO3– contents in the stream during the summer, especially in 1996, 1997, and 1998 (Fig. 6). Grimaldi and Chaplot (2000) demonstrated the lack of nitrate consumption in the watercourse itself along its entire length.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 8. Plot of NO3– versus SO42– showing variation in stream water samples based on twice-monthly monitoring between 1995 and 2000, and some source water samples in April, June, and September 1995.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
In the case of a small stream flowing over a pyrite-bearing schist substrate in the Armorican Massif, we indicate the existence of confined ground water near to and under the streambed. The aquifer is composed of a compact and strongly gleyified weathered schist. Its ground water is deoxygenated, devoid of nitrate, and enriched in sulfate, resulting from a process of denitrification coupled to oxidation of the pyrite. A study of the seasonal and storm-event-related variations of hydraulic head, combined with the use of sulfate as a chemical tracer, allows us to identify under which hydrological conditions ground water inflow influences nitrate contents in the stream water (Fig. 9) .



View larger version (56K):
[in this window]
[in a new window]
  		Fig. 9. Schematic diagrams showing the different hydrological contributions to stream flow (weathered schist ground water, shallow ground water, and sources) and their effects on nitrate and sulfate levels in the stream water at each season. The dotted line represents water table elevation. The variable relative contributions of weathered schist and shallow ground waters to the stream flow are symbolized by variable thicknesses of black arrows. The size of NO3– and SO42– symbols varies with nitrate and sulfate levels in the stream water at each season.

 
In winter, although the weathered schist ground water head is at a maximum level, its influence on nitrate contents is not visible because nitrate-rich soil waters dominate inflows to the stream. During winter stormflows, the ground water head increases still further, but the input from surface water also increases. Consequently, the nitrate contents fall especially because of dilution by surface runoff.

From winter through to summer, the level of the weathered schist ground water falls, until it drops beneath the stream bed, while the shallow ground water inflow also gradually disappears. In summer, a third hydrological compartment alone continues to feed the headwaters, corresponding to the sources of the stream. Although not zero, the nitrate contents of such sources are low in summer because of consumption by buffer vegetation around the sources. During rainy episodes in spring and summer, the weathered schist ground water level rises once again. This rise is associated with a momentary drop in the nitrate contents of the stream water.

In autumn, with the resumption of rains, the hydraulic head of the weathered schist ground water increases and becomes positive again compared with the stream bed. The relative importance of this inflow in comparison with streamflow is then clearly reflected in the measured high contents of sulfate. It is also at this time that we find the lowest contents of nitrate.

The effect of the weathered schist ground water on the chemical quality of the stream water, in particular its nitrate content, thus remains relatively limited in time. Moreover, it would appear difficult to use some type of installation to amplify such an effect. On the other hand, its duration and accentuation in the late autumn depends on the timing and intensity of the shallow ground water flow arriving in the stream, which is dependent on climatic conditions but also on the presence of landscape planning features such as hedges and buffer zones. In summer, apart from heavy rains, the weathered schist ground water does not influence NO3– contents either, since it no longer feeds the stream. However, in Western Europe, this season represents the maximum risk of eutrophication in the streams and at the coast. On the other hand, the sources represent a hydrological compartment that plays an essential role in supplying the stream during the summer. It would appear beneficial to study more precisely the role of wooded areas around the sources to decrease the nitrate contents in the stream at this season.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2004 33: 799-804. [Full Text]  




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grimaldi, C.
Right arrow Articles by Rouault, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grimaldi, C.
Right arrow Articles by Rouault, F.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Grimaldi, C.
Right arrow Articles by Rouault, F.
Related Collections
Right arrow Ground Water Quality
Right arrow Surface Water Quality
Right arrow Water Quality
Right arrow Watershed and Landscape Processes
Right arrow Nutrients

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal