SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 20 September 2006
Published in Soil Sci Soc Am J 70:1975-1982 (2006)
DOI: 10.2136/sssaj2005.0316
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
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 Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Litaor, M. I.
Right arrow Articles by Shenker, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Litaor, M. I.
Right arrow Articles by Shenker, M.
Agricola
Right arrow Articles by Litaor, M. I.
Right arrow Articles by Shenker, M.
Related Collections
Right arrow Wetland Soils
Right arrow Field-Scale Studies
Right arrow Soil Hydrology
Right arrow Phosphorus

Wetland Soils

Hydrological Control of Phosphorus Mobility in Altered Wetland Soils

M. Iggy Litaora,*, G. Eshelb, O. Reichmannb and M. Shenkerb

a Tel-Hai College, Dep. of Biotechnology and Environmental Sciences, Upper Galilee 12210, Israel
b The Hebrew Univ. of Jerusalem, P.O. Box 12, Rehovot 76100, Israel

* Corresponding author (litaori{at}telhai.ac.il)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Ground water transport of P from the altered wetland soils of the Hula Valley may influence the water quality of Lake Kinneret, which provides up to 30% of the potable water for the state of Israel. We hypothesized that land use change in these altered wetland soils is the cause for the reported increase in particulate P loading to the Jordan River, which empties to Lake Kinneret. To test this hypothesis we evaluated the P mobility from the Hula's wetland soils to waterways under various hydrological conditions using a field-scale experiment (0.8 km2) with well-monitored boundary conditions. The spatiotemporal changes in hydraulic head across the study area were measured using automated monitoring stations installed in eight observation wells. The hydraulic conductivity (K) of the near-surface peat/marl layers (179 m d–1) was significantly larger than the K values (0.001 to 0.03 m d–1) measured in the marl layers at 5 to 15 m below surface. The large K values of the near-surface layers result from the drainage of this wetland in the late 1950s followed by the oxidation of the peat layers and dissection of the lower peat and upper silt-clayey marl layers. Using a simple water-budget approach, the large field experiment yielded a discharge of 0.27 Mm3 and a loading of 306 kg P transported from the altered peat soils to the waterways, in just 7 wk. Most of the P loading was in the form of particulate P (>0.45 µm) rather than dissolved P.

Abbreviations: K, Hydraulic conductivity • RJR, Reconstructed Jordan River • SRP, soluble reactive P • TDP, total dissolved P


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE CONVERSION OF WETLANDS for agricultural use is a worldwide practice that has had a dramatic impact on water quality and quantity, both within the wetlands and downstream. For example, alteration of parts of the Everglades in the USA for agriculture has had a profound influence on the Everglades in terms of both water quality and ecology, as well as on the near-shore coral reef systems into which the Everglades drain (Davis and Ogden, 1994). Alteration of the wetlands in the San Joaquin-Sacramento Delta, USA, has had a large impact on water quality in the neighboring San Francisco Bay (Deverel et al., 1998). Similar phenomena have been observed throughout Europe (Tunney et al., 1997). Many of the water-quality changes stemming from wetland alteration are a function of the transformation of the peatlands from nutrient sinks that efficiently process and store N and P to nutrient sources that increase the potential eutrophication in downstream lakes (Richardson, 1985; Gale et al., 1995; Sharpley and Rekolainen, 1997).

The transport mechanisms of P from agricultural fields to surface waters have been studied for many years (Hart et al. [2004]; and see Correll [1998], Daniel et al. [1998], and Sims et al. [1998] for comprehensive reviews). Most of this research has focused on the loss of P due to surface runoff through soil and sediment transport. In the past, it was generally thought that subsurface flow was a minor component of P loss from agricultural drainage systems (Sharpley et al., 1994). This view is incorrect in wetland soils, where preferential flow forms during the drying period followed by rewetting, a process which affects the redox potential and reduces the sorption capacity of peat soils (Richardson and Vaithiyanathan, 1995). These hydro-geochemical changes can result in an increase in P in shallow ground water flowing from farmed peat soils to waterways (Sikora and Giordano, 1995; Gachter et al., 1998; Sims et al., 1998; Djodjic et al., 2004).

The draining of Lake Hula in the mid-1950s and the elimination of its surrounding swamps about 20 km upstream of the Jordan River inflow to Lake Kinneret, to increase the arable land in northern Israel, has had significant agricultural, ecological, and social implications. The pedological alteration has included rapid oxidation of organic matter, internal conflagration of the peat soils at shallow depth, cracking and fissure formation in the top layers (1–5 m), increased topsoil erosion by wind, significant land subsidence (up to 0.15 m yr–1), unmanageable near-surface ground water, and increased soil salinity. To reverse some of these negative effects, a small 100-ha lake (Agmon) was constructed in 1994, covering the least agriculturally productive peat soils in the Hula Valley (Fig. 1 ). The creation of the new lake elevated the water table in the center of the Hula Valley by at least 60 cm (Tsipris and Meron, 1998) and transformed the relatively dry oxidized peat soils into wet and anaerobic soil environs. Current management protocol for the valley calls for a relatively high ground water level, which will ensure that the oxidized peat soils remain wet indefinitely. Although P is the most important nutrient in regulating algal growth in downstream Lake Kinneret (Serruya and Berman, 1976), our understanding of its fate and transport from these altered Mediterranean wetland soils to waterways is inadequate. The issue is of great regional importance, since Lake Kinneret provides 30% of Israel's drinking water.


Figure 1
View larger version (56K):
[in this window]
[in a new window]
  		Fig. 1. General location of the study area (a) including a historical superposition showing the drained Hula lake and swamps (drained in the 1950s) and the recent drainage canals and new lake Agmon (b). The large field experiment is located between the reconstructed Jordan River (RJR) and Drainage Canal 303 (c).

 
Long-term monitoring of nutrient loads from the mid 1970s to the late 1990s in the Jordan River, which drains the Hula Valley, found that the valley had contributed about 10.4 Mg yr–1 of soluble P between 1993 and 1997, which was 2 Mg yr–1 more than the period before the construction of Lake Agmon (Rom, 1999). In the latter work, Rom also reported a significant increase in particulate P, from 5 Mg yr–1 before the construction of Lake Agmon to 10 Mg yr–1, afterward. On the basis of Rom's report, we hypothesized that the construction of Lake Agmon and the elevated ground water level had rewetted the previously oxidized peat soils and caused Fe-oxide dissolution with subsequent desorption of P, which could partly explain the increased P loading observed in the Jordan River. We tested this hypothesis in a series of experiments and found that the rewetting of the altered wetland soils around Lake Agmon had decreased the redox potential from greater than +400 mV to less than –220 mV (Litaor et al., 2004) and indeed increased the sesquioxide dissolution, which in turn released adsorbed and occluded P (Shenker et al., 2005). However, most of the desorbed P is most likely retained in the soils by gypsum- and calcite-originated Ca, leading to Ca-P precipitation (Litaor et al., 2005). The above-cited reports focus on various chemical processes that could account for the reported increase in P loading in the Jordan River, but none examine the hydrological control of the fate and transport of P from the altered wetland soils to waterways, which might in part occur via the well-developed macropores observed in most of the studied soils (Litaor et al., 2003). In the present study, we hypothesized that since runoff P is negligible in this altered wetland ecosystem due to high levees, most of the P transport must be governed by preferential flow via macropores. We further hypothesized that since many of the macropores are large enough to allow high ground water discharge (Fig. 2 ), particulate P could be an important mode of P transport from this altered wetland soil to waterways. The objective of the study was to evaluate the potential of P mobility in wetland soils dominated by preferential flow under various hydrological conditions using a field-scale experiment with well-measured boundary conditions.


Figure 2
View larger version (208K):
[in this window]
[in a new window]
  		Fig. 2. A pit dug in an altered wetland (Histosol) where mega macropore flow (preferential flow) has developed. This image and caption were featured on the cover of Journal of Environmental Quality, volume 32, issue 1.

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Site
The Hula Valley is the northernmost segment of the Jordan-Arava rift valley, and is approximately 70 m above sea level. It covers approximately 175 km2 and is currently drained by a system of artificial canals, which empty into the Jordan River at the southern end of the valley (Fig. 1). The climate of the Hula Valley is typical East-Mediterranean, exhibiting hot, dry summers with an average temperature of 28°C in July, and cool, wet winters with an average temperature of 12°C in January. The average rain depth in the study area is 400 mm yr–1, and the monthly evapotranspiration rate varies between 90 mm in the winter and 240 mm in the summer (Tsipris and Meron, 1998). The Hula Valley contributes an average 85 Mm3 to the overall water budget of 650 Mm3 calculated for the Jordan River (Rom, 1999). However, the valley is the main source of nutrient (N and P) loading into the river, thus any land-use change may directly impact the quality of the water.

The peat soils of the Hula Valley are predominantly Histosols (~1860 ha). Further classification on the basis of decomposition and the occurrence and quantity of CaCO3 identifies them as Medifibrists, Medihemists, Medisaprists, and ‘Conflagrated Histosols’ without lime, with minimal lime, and with lime (Department of Agriculture, 1986). Following the drainage of Lake Hula and its surrounding wetlands in the 1950s, some of the Histosols without lime exhibited highly acidic soil environs (pH < 4.0); presently, all the Histosols exhibit pHs between 5.0 and 8.0. The high organic matter content of 50 to 70% observed in 1945 declined steadily after the reclamation of the wetlands, to 30 to 50% in the higher layers of the drained peat in 1970 and 25 to 35% in 1985. A further decline in organic matter is still being observed today (Litaor et al., 2003).

To test our hypothesis, we selected an area west of Lake Agmon, which consists of shallow peat soils (<1 m) overlying a marl layer. This area was selected because we observed large macropores (>2 cm) in all soil pits excavated in the area and in more than 50 boreholes we drilled across a west-east transect. During the borehole operation, we noticed that the depth of the macropores begins at 55 to 60 cm below the surface because in each borehole, we encountered a free fall of 10 to 40 cm once the auger penetrated the aforementioned depth. Additionally, the selected area had been set aside by the regional farming authority for various ecological activities, which allowed us to manipulate the water level in the waterways around it without interfering with the intensive farming operation elsewhere in the valley.

Data Acquisition
Soils
Three soil cores of the shallow peat and underlying marl layer were hand-augured to a depth of 1.5 m in the center of the study area and each intact soil core was divided into 20-cm increments. The samples were air-dried and ground to pass through a 2.0-mm sieve while selected samples were ground to pass through a 0.5-mm sieve. The soil pH was determined in 1:10 soil/water suspensions using a glass electrode, the organic matter content was determined by the loss on ignition method (Nelson and Sommers, 1982), and the CaCO3 content was determined by a calcimeter (Nelson, 1982). The total P was extracted using the perchloric acid digestion described by Olsen and Sommers (1982). The total P was determined colorimetrically using a Spectronic Model 20 Genesis spectrophotometer (Thermo Spectronic, Rochester, NY) following Murphy and Riley's (1962) method. The extractable Fe was determined by the citrate–bicarbonate–dithionite (CBD) method described by Jackson et al. (1986) and by inductively coupled plasma–atomic emission spectroscopy (ICP–AES, Spectro, Germany). All analyses were performed in triplicate. The particle-size distribution of the mineral fraction of one core taken from a 15-m borehole, drilled in the study area (Station 1, Fig. 1) at 1-m intervals, was conducted using the pipette method (Sheldrick and Wang, 1993). The 15-m core consisted of the top peat layer and the underlying marl layers.

Shallow Ground Water
The shallow soil-aquifer system of the study area was divided into three major monitoring intervals from 2 to 5 m, 7 to 10 m, and 12 to 15 m below the surface. These monitoring depths represent two hydrostratigraphic units; the first consists of peat/marl layer (0–5 m) and the second is clayey marl layer (5–15 m). A cluster of three observation wells, (Station 1, Fig. 1) were installed 2 to 3 m apart in August 2000, enabling an evaluation of the current magnitude of the water's vertical flux suggested by Neuman and Dasberg (1977). The wells were drilled to the prescribed depth by a spiral drill (d = 15 cm) mounted on a truck. The wells were made of PVC tubes (i.d. = 9 cm) perforated by machine-sawn slots, 5 mm apart, along the three last meters of each tube and covered with nylon screen (2 mesh) to prevent clogging of the wells. The perforated sections allowed ground water flow into the wells at the depth intervals specified above. The bore hole around the perforated section was filled with coarse quartzitic sand while the rest of the bore hole was filled with swelling clay to prevent any vertical flow along the well casing. The first top meter above the ground of each well was protected by a barrel filled with cement to prevent surface disturbance. In addition, we installed eight shallow wells (0–5 m) across the study area (Fig. 1). The wells were conditioned by pumping copious amount of ground water using 3 to 4 pumping cycles. All the shallow wells in the study area were characterized by excellent hydraulics of at least 3.6 m3 h–1, which ensured that the residence time of the water in the well is very short and the samples taken from these wells represent the chemistry of this fast moving ground water system.

The hydraulic head, temperature, and electrical conductivity (EC) at Station 1 were monitored continuously from July 2001 to June 2003, using an automated system. The system included a temperature probe (APAQ-HRF, Inor, Sweden), an EC probe (WQ301), and a water-level probe (WL300 Global Water Instrumentation Inc., Gold River, CA). The water level, temperature, and EC were measured every 15 min and collected by a data logger (Hobo 8k, Onset Computer Corp., Bourne, MA). The spatial distribution of the hydraulic head was measured in eight shallow observation wells (0–5 m) located such that the changes in the ground water gradient across the entire study area could be accurately evaluated. The hydraulic heads in all eight shallow observation wells were measured continuously from 15 Aug. 2004 till the end of the experiment.

We used point-dilution technique to measure the lateral hydraulic conductivity (K) of the near-surface saturated peat layer at depths of 2.5 and 4 m using the guidelines outlined by Ward et al. (1998) and Gafni and Brooks (1990). The point-dilution method calls for the addition of a known saline solution (NaCl) to the observation well and continuously monitoring the change in EC with time. The EC of the saline NaCl solution was about 100 to 130 mS cm–1, which is 25 to 40 times higher than the background EC values in the ground water. The rate of salt dilution was measured with an EC sensor (TetraCon, WTW-25) with an operational range of 0 to 200 mS cm–1. The hydraulic conductivity at depths of 8 to 10 m and 12 to 15 m was determined using slug tests (Butler, 1998). On the basis of the particle-size analysis and field examination of the core's appearance from these depths, we assumed that the lateral and the vertical hydraulic conductivities are equal (K = Kz). The determination of hydraulic conductivity in the near-surface saturated peat layer was repeated four times, and those at 8 to 10 m and 12 to 15 m were repeated six and seven times, respectively. The water levels in the drainage canals around the study site were monitored continuously using an automated water-level gauge connected to a data logger (Thalimedes, Ott Messtechnik GMBH & Co, Kempten, Germany) from the summer of 2002 to the end of the field experiment in November 2004.

Water-Budget Calculations
The water budget of the area that served as the large field experiment, located between the reconstructed Jordan River (RJR) and drainage Canal 303 (Fig. 1 and 3) , was computed using a general discharge equation for an aquifer (Freeze and Cherry, 1979):

Formula 1[1]
where Qp is the daily precipitation, QET is the daily evapotranspiration, QVr is the artesian discharge typical to this area, QHor is the lateral discharge, and dSy/dt is the daily change in specific yield. All discharge dimensions are L3 T–1. In practice, Eq. [1] is computed using the water flux across a unit area:

Formula 2[2]
where, JP is the precipitation flux, JET is the evapotranspiration flux, JVr is vertical flux, JHor is the lateral flux, dHGW/dt is the daily rate of change in the hydraulic head (H) in the top layer, and ne is the effective porosity of that layer. All flux units and the dHGW/dt are in m d–1. The JP and JET were routinely measured in the study area (Tsipris et al., 2003). The vertical flux (JVr) was computed from the measured heads at 5 and 10 m and the vertical hydraulic conductivity (Kz) using Darcy's law:

Formula 3[3]
where dH(5–10) is the hydraulic-head difference between the heads measured at depths of 5 and 10 m and dlVr equals 4 m, which is the vertical distance between the depth where most of the macropores end (~3 m below surface) and the depth of the top opening in the 10-m observation, which was firmly installed in the second hydrostratigraphic unit. Combining Eq. [2] and [3] yielded the lateral flux equation:

Formula 4[4]
A positive lateral flux represents the daily recharge from the RJR to the shallow aquifer, while a negative lateral flux represents the daily discharge from the aquifer to the drainage canal.


Figure 3
View larger version (75K):
[in this window]
[in a new window]
  		Fig. 3. A cross-section of the large field experiment with a schematic representation of all the computed and measured flow components. The JP is the precipitation flux, JET is the evapotranspiration flux, JVr is the vertical flux, JHor is the lateral flux, dHGW is the daily rate of change in the hydraulic head (H) in the top layer, the dH(5–10) is the hydraulic-head difference between 5 and 10 m depth and dl is the vertical distance between the perforated sections of the wells.

 
Large Field Experiment
We used these simplified water-budget calculations in a large field experiment during the dry season (autumn of 2004). Hence, the potential P transport was only a function of changing the boundary condition by raising the water level in the RJR which caused ground water lateral flow, as schematically illustrated in Fig. 3. During this experiment there was no infiltration flux coupled with the change in redox potential that occurs during the typical winter wetting of the top oxidized soils (Litaor et al., 2005). The large field experiment was performed in two phases: first, we raised the RJR level by 15 cm in the upstream control gate (RJR-UP) commencing 5 Sept. 2004, while monitoring the change in hydraulic head in real time, in situ at eight observation wells (Fig. 1c), as well as the water level in the receiving Canal 303. Second, we raised the RJR water level in the downstream control gate (RJR-DOWN) on 27 Sept. 2004, while monitoring the change in the hydraulic head across the entire experimental field and monitoring the water rise in Canal 303. The ground water in the observation wells was sampled and the concentrations of soluble reactive P (SRP), total dissolved P (TDP), and total P were determined using standard methods (Murphy and Riley, 1962; Standard Methods, 2000). All P analyses were conducted using a Spectronic Model 20 Genesis spectrophotometer equipped with a 5-cm cuvette that facilitated a minimal detection limit of 5 µg L–1. The analyses were done in triplicates using appropriate standard range for calibration. No purging of the wells was conducted during the sampling campaigns because of negligible drawdown during pumping and purging trials. The difference between total P and TDP represents the particulate P fraction. The locations of the observation wells and the length of the experiment were designed to provide a reliable spatiotemporal distribution of the various P fractions. We calculated P loading from the soils to waterways using the averaged P concentrations measured during the experiment, multiplied by the estimated water discharge.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
General Hydrological Characterization
The soils of the study area were formed along a stratigraphic/pedologic boundary zone between the old Lake Hula and the swamps to the north (Fig. 1). However, the spatial distribution of this boundary is quite fuzzy, because of seasonal migration of the swamps/lake boundary. Hence, the soils in the study area are characterized by highly rich calcareous horizons (>600 g kg–1) at shallow depth (60 cm +), covered by the altered peat layers from the old swamps, which exhibit high organic matter content (Table 1). The layers exhibiting the highest organic matter content (>400 g kg–1) are also modestly acidic, reminiscent of the acidic conditions that prevailed in the swamps before the drainage. A sharp increase in pH from 5.6 to higher than 7 was observed across the depositional boundary between the altered peat horizons and the calcareous layers of the old Lake Hula. In general, higher values of extractable Fe and total P were observed in the peat layers relative to the calcareous horizons (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Mean and standard deviation of selected soil attributes measured from the three cores augured to a depth of 1.5 m. This depth represents the shallow peat layer and underlying marl layer.

 
The silt content is the main particle-size fraction, averaging 65% throughout the two hydrostratigraphic units to a depth of 15 m, followed by a mean clay content of 27%, whereas the mean sand content is 8% and disappears completely at a depth of 13 m below the surface. The vertical distribution of the hydraulic conductivity showed that the soil at the 2.5-m depth is highly conductive with K values of 179 m d–1, whereas the deeper layers are significantly less conductive, exhibiting K values of 0.03 to 0.001 m d–1 (Table 2). During the point-dilution tests, we observed a significant decrease in EC values at a depth of 2.5 m within 10 h of the start of the experiment, while the EC values at a depth of 4 m did not change at all. This finding suggests that most of the macropores are limited to a depth no greater than 3 to 3.5 m below the surface.


View this table:
[in this window]
[in a new window]
  		Table 2. Geometric mean and standard deviation of K values measured at three depths.

 
The K values in the lower section of the profile were more in agreement with the expected K values for a silt-clayey texture (Freeze and Cherry, 1979). The K values of the uppermost layer were at least three orders of magnitude higher than those expected from the particle-size distribution (0.1 m d–1). These high K values most likely resulted from the numerous cracks and fissures originating from the drainage of the wetlands followed by rapid oxidation of the organic matter and shrinkage of the silt-clayey layer underneath. To further examine this interpretation, we opened over 20 pits (depth > 5 m) in the area around Lake Agmon where the peat layer is shallow (<1 m) and overlying a marl layer, as deduced from the regional soil map (Department of Agriculture, 1986). We found large cracks and fissures of several centimeters in width (Fig. 4 ) in all excavated pits, which explained the unexpectedly large K values. We also opened 10 pits in the area north of Lake Agmon, which is known for its deep peat layers and absence of marl layer in the top 15 m of the shallow soil/aquifer system. The deep peat layers were never exposed to the atmosphere, thus we found no cracks or fissures, explaining the low K values measured in this location (0.001 m d–1). The large K values determined in the present study have also been reported in other wetland ecosystems (Mitsch and Gosselink, 2000) and they are usually attributed to the repeated cycles of drying and rewetting of the organic-rich wetland soils (Holden and Burt, 2002, 2003).


Figure 4
View larger version (138K):
[in this window]
[in a new window]
  		Fig. 4. Typical cracks and fissures observed in the peat layers during the pit survey. Similar macropores were observed in all of the excavated pits.

 
The temporal hydraulic-head distribution in this altered peat soil system (Fig. 5 ) suggested that the lowest hydraulic head is always observed at the end of the summer of each monitoring year, and it begins to rise well before the rainy season (December through April), in response to a rise in water level in the RJR. The water level in the RJR is manually controlled to accommodate the ecological needs of Lake Agmon, as well as irrigation requirements across the study area. Consequently, the rising water in the RJR generated an increase in the ground water level, even during late summer and early autumn, when there is no rain or other source of recharge. The temporal changes in the hydraulic heads of all measured depths were fairly similar and did not exceed 60 cm, except during the exceptionally wet winter of 2003, which caused wide flooding throughout the valley. The modest fluctuations in hydraulic head throughout the monitoring years suggest continuous recharge from the RJR. This recharge mechanism is quite efficient because it greatly minimized the influence of the strong ET flux, especially in the summer (210 mm per month), as well as the potential artesian flow component suggested by Neuman and Dasberg (1977). In fact, the vertical flux (Jvr) was reduced to nil in response to a rise in water level in the RJR. During the rainy season, the ground water recharge was quick, followed by a rapid decline in ground water level as the storm ceased. The quick response of the ground water level to aerial recharge is additional evidence of the effectiveness of macropores in transferring water flow in these shallow peat soils.


Figure 5
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 5. Temporal distribution of hydraulic heads (m) measured at three measuring depths at station 1.

 
Water-Budget Calculations
The large field experiment consisted of elevating the water level in the RJR and monitoring the changes in the hydraulic heads over time (Fig. 6 ). During the first phase of this experiment, the wells in the northern part of the study area responded rather quickly and rose intermittently, depending on their location (from 5 to 26 cm in 2 wk, for brevity only well 403/4 is shown in Fig. 6). Concurrently, the water level in the upstream segment of drainage Canal 303, which collects the ground water discharge, rose by 13 cm. On the other hand, the wells in the southern section and the downstream segment of drainage Canal 303 (disconnected from the northern end by an earthen dam) did not respond at all to the rise in water level in the upstream section of the RJR. Three weeks into the experiment, we reversed the recharge boundary conditions by opening the wooden control gate to lower the water level in the upstream section. At the same time, we raised the water level at the southern end of the RJR by 60 cm, which created local flooding. We readjusted the height of the wooden control gate and as a result, the water level dropped by 18 cm and remained at that level for 4 wk (Fig. 6). The rise in the water level at the southern end of the RJR was followed by a sharp increase in hydraulic head (42 cm in Station 1[Fig. 6] and 45 cm in Well 403/1, not shown) in the southern part of the field in <3 wk, which eventually led to a measurable rise in the water level at the southern end of drainage Canal 303. Analysis of the propagated increase in hydraulic head in time and space in the large field experiment suggested that the flow direction is mainly from east to west, which is perpendicular to the regional hydraulic gradient (Tsipris and Meron, 1998). This evidence suggests that the preferential flow also dictates the direction of the water flux.


Figure 6
View larger version (30K):
[in this window]
[in a new window]
  		Fig. 6. The hydraulic head changes and water levels in RJR and the 303 drainage canal (cm) during the large field experiment. The symbols Up and Down represent the automatic water level measurements at the top and the bottom sections of each waterway. Water change in the Down station of the 303 canal was registered only after 26 d of the experiment.

 
We entered the data of the large field experiment into Eq. [4], along with the measured hydraulic-head difference at each observation well, measured evapotranspiration, and estimated effective porosity (~0.25, Michalson, 1992). Solving Eq. [4] yielded the horizontal flux summarized in Table 3. The discharge for the entire large field experiment was calculated by multiplying the average horizontal flux by the area of the upper and lower zones respectively (Table 3). These discharge calculations are quite conservative, because we chose the lowest published estimation of effective porosity.


View this table:
[in this window]
[in a new window]
  		Table 3. Horizontal flux, discharge and P concentrations and loadings during 7 wk of the large field experiment. The total discharge was calculated by multiplying the average horizontal flux by the area of the upper and lower zones, respectively. The total P flux was calculated by multiplying the average P concentrations by the total discharge per zone.

 
We estimated the three forms of P flux in the large field experiment by multiplying the discharge (0.27 Mm3 per period) by the means of each form of P as measured in the wells of the studied area during the experiment. As expected, the SRP exhibited the lowest P flux, followed by the TDP flux, whereas the total P exceeded 300 kg for the period of monitoring. The large field experiment demonstrated that ground water flow through macropores can yield substantial P loading from the altered peat soils to waterways, and that most of the P loading comes in the form of particulate, rather than dissolved P. We attributed this P-fractionation pattern to the prevalent occurrence of large macropores, which facilitated relatively fast flow conditions, as was demonstrated in the large field experiment. The fast flow of the shallow ground water is probably the mechanism of particle detachment, with the subsequent increase in particulate P.

The large P flux generated by the large field experiment supports the hypothesis advanced by Simard et al. (2000) that preferential flow pathways are important conduits that transport large amounts of P from diverse soil systems under various land-use practices to waterways. The fact that most of the transported P was in the form of particulate P supports Hart et al.'s (2004) statement that the form of P transport in subsurface and surface flow is highly site-specific and no generalization can be made regarding the form of P loss via subsurface flow. Moreover, the form of P transport in subsurface flow at a given site is also controlled by the specific initial conditions per event. For example, if the large field experiment had been conducted during the wet season, the total amount of P would have increased, due to P translocation from the top soil layers. Under wet conditions, the form of P would have changed, since more flow would have come from the soil matrix where the change in redox potential has a greater impact on P desorption because of increased sesquioxides dissolution under reducing conditions. However, since the experiment was conducted in the dry season, the top soil layers (0–100 cm) remained dry with no change in redox conditions, thus no P was mobilized downward.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The main conclusions drawn from this study are that: (i) significant ground water discharge (0.27 Mm3 per 7-wk period) via the cracks and fissures of the altered wetland soils can mobilize considerable P loading (>300 kg P) into waterways, (ii) most of the loaded P is in the form of particulate P (>0.45 µm) rather than dissolved P, (iii) proper water management of the RJR and other major drainage canals may minimize P transport from the altered wetland soils to waterways and (iv) the observed increased of particulate P in the Jordan River resulted in part by inadequate water management of this altered ecosystem.


   ACKNOWLEDGMENTS
 
This research was supported in part by the EU project PROWATER, EVK1-CT1999-00036, the Israeli Water Commission, and by the GLOWA- Jordan River Project funded by the German Ministry of Science and Education (BMBF), in collaboration with the Israeli Ministry of Science and Technology (MOST).

Received for publication September 25, 2005.


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




This article has been cited by other articles:


Home page
 Soil Sci.Home page
S. Brand-Klibanski, M. I. Litaor, and M. Shenker
Overestimation of Phosphorus Adsorption Capacity in Reduced Soils: An Artifact of Typical Batch Adsorption Experiments
Soil Sci. Soc. Am. J., June 8, 2007; 71(4): 1128 - 1136.
[Abstract] [Full Text] [PDF]

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 Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Litaor, M. I.
Right arrow Articles by Shenker, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Litaor, M. I.
Right arrow Articles by Shenker, M.
Agricola
Right arrow Articles by Litaor, M. I.
Right arrow Articles by Shenker, M.
Related Collections
Right arrow Wetland Soils
Right arrow Field-Scale Studies
Right arrow Soil Hydrology
Right arrow Phosphorus

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