JEQ Journal of Natural Resources and Life Sciences Education
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 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 (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bernal, S.
Right arrow Articles by Sabater, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bernal, S.
Right arrow Articles by Sabater, S.
Agricola
Right arrow Articles by Bernal, S.
Right arrow Articles by Sabater, S.
Related Collections
Right arrow Ecosystem Management
Right arrow Nitrogen
Right arrow Plant and Soil Interactions
Right arrow Nutrient Cycling
Journal of Environmental Quality 32:191-197 (2003)
© 2003 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORTS
Landscape and Watershed Processes

Leaf Litter Dynamics and Nitrous Oxide Emission in a Mediterranean Riparian Forest

Implications for Soil Nitrogen Dynamics

S. Bernal*, A. Butturini, E. Nin, F. Sabater and S. Sabater

Departamento Ecología, Universidad de Barcelona, Avd. Diagonal 645, 08028 Barcelona, Spain

* Corresponding author (sbernal{at}porthos.bio.ub.es)

Received for publication December 18, 2001.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SITE DESCRIPTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Mediterranean riparian zones can experience severe drought periods that lead to low soil moisture content, which dramatically affects their performance as nitrate removal systems. In the Mediterranean riparian zone of this study, we determined that N2O emission was practically nil. To understand the role of forest floor processes in nitrogen retention of a Mediterranean riparian area, we studied leaf litter dynamics of two tree species, London planetree [Platanus x acerifolia (Aiton) Willd.] and alder [Alnus glutinosa (L.) Gaertn.], for two years, along with soil nitrogen mineralization rates. Annual leaf litter fall equaled 562.6 ± 10.1 (standard error) g dry wt. m-2, 68% of which was planetree and 32% of which was alder. The temporal distribution of litterfall showed a two-peak annual cycle, one occurring in midsummer, the other in autumn. Planetree provided the major input of organic nitrogen to the forest floor, and the amount of planetree leaves remaining on the forest floor was equivalent to approximately four years of stock. Leaf litter decomposition was three times higher for alder (decay coefficient [k] = 1.13 yr-1) than for planetree (k = 0.365 yr-1). Mineralization rates showed a seasonal pattern, with the maximum rate in summer (1.92 mg N kg-1 d-1). Although the forest floor was an important sink for nitrogen due to planetree leaf accumulation, 7.5% of this leaf litter was scoured to the streambed by wind. This loss was irrelevant for alder leaves. Due to the litter quality, the forest floor of this Mediterranean riparian forest acts as a nitrogen sink.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SITE DESCRIPTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
MANY studies have analyzed nitrogen dynamics in riparian ecosystems to learn which mechanisms are involved in removing nitrogen from surface and subsurface waters in those ecosystems (see Hill, 1996, for a thorough review). Vegetation uptake and denitrification appear as the two major biological processes involved in N export in riparian areas (Lowrance et al., 1984; Pinay and Decamps, 1988; Schnabel et al., 1997; Verchot et al., 1997). However, several environmental and biogeochemical conditions limit those processes: vegetation uptake is limited by light and nutrients, while denitrification is limited by soil water saturation and nitrate and organic carbon abundance (Knowles, 1981). A recently concluded European project (NICOLAS ENV4-CT97-0395) focused on the potential of the riparian systems to remove nitrogen through a climatic gradient. In arid and semiarid riparian areas, such as those of the Mediterranean regions, the soil is not water saturated and ground water flows several meters below the soil surface (Butturini et al., 2002). Consequently, in a Mediterranean riparian system, the superficial fluxes occurring at the soil organic layers and the subsurface fluxes should be accounted separately. Concerning the superficial fluxes, litter dynamics is an important factor to determine whether the pool of particulate organic matter in the forest floor may act as a source or sink for nitrogen (Greenway, 1994). Several studies have coincided in pointing to the importance of leaves in relation to the total litter (leaf litter = approximately 80% of the total litter) (Meentemeyer et al., 1982; Sharma and Ambasht, 1987; Cañellas and San Miguel, 1997; Gallardo et al., 1998). Furthermore, although there are studies on the effect of climatic factors on the distribution and transfer of mineral nutrients to the soil through litterfall (Bray and Gorham, 1964), little is known about how strongly dry periods may affect leaf litter dynamics and further, the consequences on nitrogen retention capacity of the system. On the other hand, studies on denitrification conducted in arid and semiarid areas concluded that when soil moisture content is well below field capacity, N2O fluxes are low (Mummey et al., 1994; Jorgensen et al., 1998). No previous studies on how the conjunction of those mechanisms may affect the nitrogen retention capacity of riparian soils under Mediterranean climate seem available. This paper develops this approach in a Mediterranean riparian forest composed of two main tree species characterized by a very different litter quality: alder, with soft, N-rich leaves (Quinn et al., 2000), and planetree, a very common species used in reforestation, with a very low nutritional value (Malicky, 1990). The riparian forest is located in a Mediterranean area in Catalonia, in northeastern Spain, which experiences severe summer drought. The study site is not receiving important inputs of nitrogen from agricultural activities via soil surface flow and therefore, the potential of this forest soil in nitrogen retention can be assessed without pollution interference.

This study aims to determine when the forest soil of a Mediterranean riparian forest acts as a nitrogen source or sink, and whether seasonal variation may affect this behavior. To accomplish this, the study is focused on the following objectives: (i) monitor the litterfall of the two described species during two vegetative cycles to quantify leaf litter production and at the same time determine temporal and spatial distribution; (ii) determine the differential effects of the two tree species on nitrogen forest floor dynamics from the study of their leaf decomposition rates; and (iii) estimate the N2O emission rate to determine the significance of this process in the exportation of nitrogen from the system. In addition, the N input from bulk rainfall and the N content in surface overland flow have been studied, as well as nitrogen mineralization and the most relevant soil characteristics, to generate an overall picture showing the inputs and the outputs from the studied riparian forest.


   SITE DESCRIPTION
TOP
 ABSTRACT
 INTRODUCTION
 SITE DESCRIPTION
MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The study site, Fuirosos, is located in a forested catchment (16.2 km2) with an altitude range between 100 and 770 m above mean sea level, 65 km from Barcelona, in northeastern Spain (41°42' N, 2°34' W). Climate is typically Mediterranean, with a mean annual temperature ranging from 6.9°C, in January, to 24°C, in August (winter air temperatures seldom drop below 0°) (Fig. 1) . Precipitation is distributed irregularly (number of days does not exceed 70 per year), autumn and spring having the most rain while storms occur only occasionally in summer (Fig. 1). The riparian area is crossed by an intermittent third-order stream with a flow that usually ceases from July–August until September–October. Annual average water discharge ranges from 5 to 20 L s-1. Soils are dominated by sand and fine sand (46 and 24% respectively), with silt and clay at 15% each (Sala, 1983).



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 1. Temporal dynamics of (a) daily precipitation (mm), (b) air temperature (°C), and (c) ground water table (cm) in the Fuirosos riparian area during the study period (Butturini et al., 2002). Data unavailable is indicated by NA.

 
The study site is a forested zone, 55 m long and 18 m wide (990 m2), located between an agricultural field (1.1 ha) and the stream. The average slope across the riparian section is 21%, although the slope between the agricultural field and the riparian forest is nil. The A horizon of the soil is approximately 0.5 cm deep. The soil moisture in the riparian area averaged 17%, exceeding 25% during storms, while decreasing gradually to 10% during the drought period (Butturini et al., 2002). A small area (8 m2) had higher soil moisture than the rest of the study site during the wet season, due to the proximity of the ground water to the soil surface at that point. In this area, the soil moisture during the wet season rose to 40%. Ground water flows 1 to 2 m under soil surface during the wet season (winter and spring), while in summer the water table suffers an abrupt drop of approximately 1 m (Butturini et al., 2002), and therefore, the soil never became water saturated during the studied hydrological periods. The ground water level recuperates with the first autumn rains (Fig. 1). The vegetation in the riparian zone was dominated by a plantation of planetree with a corridor of alders flanking the stream. The riparian forest was young (the last logging was done 16 yr ago) and the average diameter of trees was 12.6 cm (planetree) and 12.2 cm (alder) (estimated in 1999). Shrubs were scarce and short, mainly limited to the external part of the riparian area, with an abundance of blackberries (Rubus ulmifolius Schott).


   MATERIAL AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SITE DESCRIPTION
 MATERIAL AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The study was conducted from June 1998 to February 2000. During the study period, samples of rainfall and surface overland flow were collected and analyzed for inorganic nitrogen forms, to determine the inorganic nitrogen inputs and outputs in the riparian area. Surface overland flow entering the riparian area from the upland was considered nil, as long as the slope at that location was nil. The input and the output of particulate organic matter as leaf litter were also monitored. To determine the fluxes of particulate organic matter to inorganic forms, soil mineralization experiments and leaf decomposition experiments for the two main tree species were performed. Finally, N2O emission was measured to highlight the proportion of nitrogen that was released directly into the atmosphere.

Water Chemistry
The atmospheric inputs of nutrients were sampled with a bulk deposition collector. After each rain event the rainwater from the collector was analyzed for inorganic forms. Surface overland water was collected with an 8-m channel connected to a polyvinyl chloride water collector. The channel collected the soil surface drainage of an 8.3-m2 area. Water samples for chemical analysis were taken each time and filtered through glass filters (Whatman [Maidstone, UK] GF/F). Both NH+4 and NO-3 were analyzed colorimetrically with a Technicon Autoanalyzer (Technicon, 1976), NH+4 with the indophenol blue method (Keeney and Nelson, 1982), and NO3- with the Griess–Ilosvay method (Keeney and Nelson, 1982) after reduction by percolation on a copperized cadmium column.

Leaf Litter Dynamics
Vertical and lateral litterfall dynamics were monitored from June 1998 to January 2000. The annual litter production, that is, the vertical input of particulate organic matter (expressed as g dry wt. m-2 d-1), was measured during two consecutive years by periodically sampling 17 plastic cylindrical litter baskets (0.13 m2) that were randomly distributed over the study site (Ovington and Murray, 1964). Each basket was constructed from 2-mm mesh and fixed to a wooden post of approximately 1 to 1.5 m from the forest floor. The annual litter that drifted down to the stream, that is, the lateral output of particulate organic matter (expressed as g dry wt. m-1 d-1), was estimated with six litter bank traps. In this case, the traps (0.7 m long) consisted of a wooden frame covered with 2-mm mesh that formed a pocket with the open side facing upslope and parallel to the stream to collect the litter transported from the riparian area to the stream. The sampling intervals for the litter baskets and bank traps varied from weekly to monthly depending on the amount of material in the traps. The leaf litter accumulation over the forest floor was estimated twice in 1998, on the 17 and 28 July, by defining several parcels in the riparian area (0.12 m2, seven replicates). After collection, leaf litter from vertical and lateral traps and from the forest floor was dried at 80°C, sorted into leaves of planetree and alder, and weighted to the nearest 0.01 g. Carbon and nitrogen contents were determined with a Carlo Erba (Milan, Italy) NA 1500 elemental analyzer.

Leaf litter decomposition was measured for planetree and alder. Freshly fallen leaves were collected during fall from individual trees, dried (70°C, 48 h), and stored. Subsequently, 150 polyethylene 0.3-mm mesh bags were filled with litter (Singh and Gupta, 1977): 100 bags were filled with 0.8 g dry wt. of planetree leaves, while 50 bags were filled with 0.6 g dry wt. of alder leaves. Ropes of six bags each were randomly distributed over the forest floor of the studied area. One bag from each rope was retrieved at Days 0, 21, 41, 83, 169, and 350 (from October 1998 to October 1999). After collection, leaves were gently cleaned to remove attached sand and soil particles, and dried (70°C, 48 h). The loss of weight was expressed with the exponential decay model Wt = Woe-kt, where Wt is the amount of material remaining at time t, W0 is the amount of material at time 0, and k is the decay coefficient (Petersen and Cummins, 1974). Exponential regression decay was considered significant at p < 0.01. Each time bags were collected, the total carbon and nitrogen content was analyzed.

Soil Characterization and Processes
The organic matter content and the organic and inorganic nitrogen compounds in the first 10 cm of soil were measured seasonally from nine replicate samples each time. The percent organic matter in the soil was determined by ignition of the soil samples (450°C, 24 h) (Rowell, 1994). The soil inorganic nitrogen compounds were determined after extraction with K2SO4 and subsequent filtration though glass filters (Whatman GF/F). Both NH4+ and NO3- were analyzed colorimetrically with a Technicon Autoanalyzer (Technicon, 1976) as described above. The organic nitrogen fraction was determined after oxidation of the soil extraction with a solution of K2S2O8, H3BO3, and NaOH (1 h, 120°C) and posterior subtraction of the nitrogen inorganic fraction. Total carbon and nitrogen content were determined with a Carlo Erba NA 1500 elemental analyzer.

Net nitrogen mineralization was calculated seasonally by the changes measured in the mineral nitrogen content of largely undisturbed soil samples placed inside polyethylene bags that allowed air to pass through them but prevented leaching. After one month of incubation in the field the inorganic nitrogen content in the bags was compared with the content at the beginning of the incubation (Ellenberg, 1977).

Nitrous oxide emission was measured seven times from October 1998 to June 1999 by installing four to six polyvinyl chloride gas chambers (20-cm diameter) driven 6 cm into the soil on each occasion. The gas chambers consisted of a rubber septum as a sample port and a small perforation as a pressure vent (Denmead, 1979). The gas flux from the soil surface was determined by sampling the headspace air of the chambers after 1 and 4 h of incubation. The gas samples were collected by syringe through a septum and analyzed for N2O with a gas chromatograph equipped with an electron capture detector (63Ni) and two Poropak Q filled columns (3.2 mm in diameter x 3.1 m long) (Astorga et al., 1993).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SITE DESCRIPTION
 MATERIAL AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Dissolved Inorganic Nitrogen in Bulk Precipitation and Surface Overland Flow
Precipitation events were irregularly distributed, concentrated mainly in autumn and spring and accounting for 520 and 449 mm for 1998 and 1999, respectively. Nitrate and ammonia concentration in precipitation ranged from 0.05 to 3.22 mg NO3–N L-1 and from 0 to 9.84 mg NH4–N L-1, respectively, with a total nitrogen input from rainfall of 411.54 mg m-2 yr-1 (45.7 and 54.3% as NO3–N and NH4–N, respectively).

There was no surface overland flow during the events that occurred from January 1999 to April 1999. Nitrate and ammonia showed a wide range of concentration in overland flow, from 0 to 14.47 mg NO3–N L-1 and from 0 to 9.63 mg NH4–N L-1, respectively. For both nitrate and ammonia, the maximum concentrations were found in the overland flow of rain events from May to August (Fig. 2) . In any case, when concentrations were converted to loads, no seasonal changes were observed, either for nitrate or ammonia. In total, surface overland flow contributed to an output of dissolved inorganic nitrogen equal to 52.35 mg m-2 yr-1 (76.1% as NH4–N and 23.9% as NO3–N).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 2. Ammonia (dark circles) and nitrate (open triangles) concentrations (mg L-1) in (a) rain and (b) surface overland flow.

 
Leaf Litter Dynamics
The period of study included two complete litterfall cycles. Total leaf litter production in 1998 and 1999 was equal to 572.73 and 552.49 g dry wt. m-2, respectively, both years with a similar species contribution: 67.4% corresponded to planetree and 32.6% to alder. However, the main part of the alder leaf litter production falls directly into the stream channel, and therefore only 26.3% in 1998 and 11.3% in 1999 of the alder leaf litter reached the forest floor. The leaf litter fall temporal distribution was characterized by a two-peak yearly cycle consisting of a first peak in midsummer and a second in early autumn. In 1998, the summer peak accounted for 23.2 and 21% of the total planetree and alder leaf litter fall, respectively, while in 1999 the relative importance of the summer peak was double, 44.15 and 50% for planetree and alder, respectively (Fig. 3) . Part of the material that had reached the forest floor was transported by the wind into the stream. This fraction consisted of 727 and 62 g dry wt. m-1 yr-1 of planetree and alder leaves, respectively. The temporal distribution of the lateral output was different for the two species, with alder leaf contribution being relatively constant throughout the study period, while planetree leaves had several maxima, mainly in winter (Fig. 3). Considering that the nitrogen content in the leaf litter of planetree was 1.65 ± 0.04 (n = 8) during the summer peak and 0.76 ± 0.03% (n = 8) (planetree C to N ratio = 62.9) in autumn, and that the nitrogen content for the alder leaves was 1.98 ± 0.11% (n = 8) (alder C to N ratio = 24.6), it has been estimated that in 1998 the annual input of coarse organic nitrogen to the forest floor as leaves accounted for 4.7 g N m-2 (79.3% planetree leaves and 20.7% alder leaves). In the same way, in 1999 the vertical input of leaves on the forest floor was equivalent to 4.69 g N m-2 (91.4% planetree leaves and 8.6% alder leaves). Planetree leaves were predominant over the forest floor, the leaf stock being equivalent to 1.38 ± 0.20 kg dry wt. m-2 (n = 13).



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 3. Daily rate (average and standard deviations) of leaf litter input via direct fall (top) and output via lateral transport (bottom) in Fuirosos during 1998 and 1999. Black circles, planetree; open circles, alder.

 
The rate of leaf litter decomposition was lower for planetree than for alder (0.36 and 1.13 yr-1, respectively). During the first year nitrogen tended to accumulate in the leaf litter: during the first 83 d of the experiment, alder leaves increased their N content to 3.4%, which represented 33% more N than at t = 0. This was followed by a decay to a percentage similar to the initial one (Fig. 4) . Planetree leaves showed a slower increase in N content than alder leaves. Between Days 0 and 169 of the experiment the increase was gradual and reached up to 33% more than the initial content. From Day 169 to the end of the experiment the percentage of N content remained more or less constant (Fig. 4). Regarding the carbon content of the leaves, the initial percentage for both species was 47.5%, and there was a gradual decrease over the year down to 10% of the initial content (Fig. 4).



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Nitrogen and carbon dynamics of leaf litter during the first year of decomposition for (a) planetree and (b) alder. For each nutrient, solid lines indicate the percentage of the initial content remaining at various intervals. Open triangles for nitrogen. Dark circles for carbon. Dashed line + open circles indicates weight loss. The solid line is the regression line for the exponential decay (for planetree: r = 0.94, df = 4, p < 0.01; for alder: r = 0.97, df = 4, p < 0.01).

 
Soil Characterization and Processes
The fraction of soil organic matter was similar in winter and spring, and increased significantly (t = 2.87, p < 0.05, n = 7) in summer, averaging 6.12 ± 0.72% (n = 7) of the dry weight (Table 1). The highest C to N ratio was estimated in winter (16.95 ± 0.62, t = 3.5, p < 0.05, n = 12) with a mineral soil nitrogen percentage of 0.15. Nitrate concentration in the soil showed a minimum in spring (1.45 ± 0.33 mg NO3–N kg-1, n = 12) and a maximum in summer (10.3 ± 0.35 mg NO3–N kg-1, t = 8.9, p < 0.05, n = 8). Similarly, organic nitrogen was maximum in summer and minimum in spring. Ammonia also followed a seasonal pattern with a maximum in summer and a minimum in winter (Table 1). Maximum mineralization rates occurred in summer and were estimated at 1.92 ± 0.2 mg N kg-1 d-1 (n = 12). In contrast, mineralization rates were significantly lower in autumn and winter (0.55 ± 0.09 mg N kg-1 d-1, n = 24, t = 6, p < 0.05, n = 12) (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Average nitrate (NO3–N), ammonia (NH4–N), and dissolved organic nitrogen (DON) concentrations in the first 10 cm of soil, and mineralization rates (MR), percent organic matter (OM), nitrogen content, and C to N ratio for each season.

 
Maximum N2O emission rates occurred at the end of the wet season (17.2 µg N m-2 d-1), while the minimum was during the dry season (0.6 ± 0.1 µg N m-2 d-1, n = 12). The delimited area with higher soil moisture (35 vs. 21%) showed a similar pattern, but the estimated values of N2O emission ranged from 68 µg N m-2 d-1 in winter to 7.16 µg N m-2 d-1 in summer.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SITE DESCRIPTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
In Fuirosos the leaf litter, mainly of planetree, provided the major input of nitrogen on the forest floor, 4.7 g N m-2 yr-1 (92% of the total input). The bulk precipitation input of inorganic N (approximately 410 mg N m-2 yr-1) accounted for the remaining fraction. Escarré et al. (1999) have reported similar values of bulk inputs of inorganic N (530 mg N m-2 yr-1) for a Mediterranean forest located in Montseny, 30 km from our study site. On the other hand, N2O emission in Fuirosos (0–25 mg N2O-N m-2 yr-1) was not a significant nitrogen export mechanism (<0.5% of the total N input), probably due to the low soil water saturation and low inorganic N content. In contrast, several authors reported higher N2O fluxes for temperate soil forests: in Germany, 490 and 40 mg N2O-N m-2 yr-1 for alder and beech (Fagus spp.) forests, respectively (Mogge et al., 1998), or 66 mg N2O-N m-2 yr-1 for sycamore (Platanus spp.) forests in Scotland (Skiba et al., 1996). In semiarid zones N2O fluxes are similar to those estimated in Fuirosos: 0 to 18 mg N2O-N m-2 yr-1 for a semiarid shrub-steppe in North America (Mummey et al. 1997), or 0 to 2 mg N2O-N m-2 yr-1 for a dry stubble field (Jorgensen et al., 1998). Nevertheless, those studies suggest that under semiarid conditions N2O fluxes may be substantial only at times following precipitation events due to the release of readily decomposed organic matter and inorganic nitrogen for nitrification after wetting of the dry soil. Thus, it should be taken into account that in Fuirosos the role of N2O emission in the removal of nitrogen may be slightly underestimated.

Removal of nitrogen via lateral transport of leaves was found to be the most important output flux (375 mg N m-2 yr-1, approximately 7.5% of the total nitrogen inputs), mainly consisting of planetree leaf litter. This fact implies that in Fuirosos 92% of the total annual nitrogen input was retained in the riparian forest floor. Campbell et al. (1992) also found that 470 g dry wt. m-2 yr-1 (10% of the total leaf litter entering to the forest floor of a forested riparian zone in southeastern Australia) was removed via lateral transport by wind blow. In Fuirosos, the surface overland flow (52.35 mg N m-2 yr-1) did not account for a substantial output of nitrogen (only 1% of the total N input).

The decomposition rates show that the alder leaves decomposed in less than one year, while the planetree leaves, with a k rate three times lower, tended to accumulate on the forest floor. The planetree C to N ratio of fresh leaf litter was equal to 62.9, which implies a low mineralization rate because a "critical" C to N ratio for mineralization seems to be between 20:1 and 30:1 (Lutz and Chandler, 1946). The increase of nitrogen in leaf litter during the first year of decomposition may be a consequence of this low mineralization, as suggested by Schlesinger and Hasey (1981). Studies in other semiarid environments have reported similar mean nitrogen mineralization rates (Zhang and Zak, 1998) (0.66 mg N kg-1 d-1). Regarding inorganic N in mineral soil, similar values had been reported for other Mediterranean areas like Prades and Montseny (10 mg N kg-1) (Serrasolses et al., 1999), and for arid lands (7.4 mg N kg-1 in the Arid Land Ecology Reserve) (Mummey et al., 1997). The increase of N percent during the first year of decomposition and the low nitrogen mineralization rate suggest that nitrogen is retained in the organic form and slowly recycled. This parsimonious mobilization of nitrogen in litter has been already recognized for Mediterranean systems in earlier studies (Bocock, 1963; Schlesinger and Hasey, 1981).

In Fuirosos, the maxima of leaching products occurred from May to July, coinciding with the summer rain events, which are usually short and intense. It was also in summer when the maximum concentrations of organic matter and nitrogen were estimated, and mineralization rates followed a similar pattern with maxima in spring and summer. In the same way, important peaks of leaching products at the beginning of autumn have also been described for other Mediterranean ecosystems due to the arrival of the first autumn rains after the summer drought, which causes the leaching of the most soluble compounds from the recently fallen litter (Terrades, 1996; Maamri et al., 1996). Thus, in Mediterranean ecosystems, the mobilization of nutrients occurs in pulses after the precipitation events, and the effect of the sporadic summer rains may be of major importance since the alternancy of dry and humid conditions in these ecosystems stimulates microbial activity (Mummey et al., 1994; Terrades, 1996). The fact that this study area was subjected to severe drought periods may explain several aspects of the observed leaf litter dynamics. For instance, the early summer peak in leaf litter fall is probably due to water restrictions (Nilsen and Orcut, 1996). This has been described for other Mediterranean climatic zones in Australia (Bunn, 1988) and South Africa (Stewart and Davies, 1990). In Fuirosos, although litterfall started early in summer and pulses of microbial activity and soil inorganic N had been detected in summer and autumn, overland flow of inorganic nitrogen compounds had not been important. The toughness of planetree leaves probably provokes a low N availability, diminishing the effect of summer and autumn rains on the release of nutrients, and also preventing nitrogen from leaching.

Finally, the conjunction of the analyzed processes may determine if Fuirosos forest floor litter acts as a sink or source of nitrogen. Studies conducted in humid regions concluded that the riparian forests act as a sink for nitrogen, even in saturated conditions, due to high rates of denitrification and storage of nitrogen in soil organic matter (Hanson et al., 1994). Studies performed elsewhere in humid regions also consider riparian areas as buffer zones with an important nitrogen retention capacity (Lowrance et al., 1984; Pinay and Decamps, 1988; Schnabel et al., 1997; Verchot et al., 1997). In contrast, in a Mediterranean system such as Fuirosos, denitrification could be considered insignificant in terms of nitrogen removal, due to the unsaturated soil water conditions. In addition, the N2O fluxes estimated in Fuirosos were very low, suggesting that nitrification is not an important process, probably due to the low soil inorganic N content that prevents leaching of inorganic N after rain events. In Fuirosos, the accumulation of litter, mainly planetree leaves, has turned out to be an important nitrogen retention pool. The toughness and the low nutritional value of planetree leaves (Malicky, 1990; Canhoto and Graça, 1996; Pereira et al., 1998) cause slow decomposition, implying a net accumulation of coarse organic nitrogen from year to year. In that sense, although this nitrogen is slowly incorporated in the biota, the forest floor may act as a nitrogen sink. In this way, Greenway (1994) described a woody riparian system in southeastern Queensland (Australia) that also acted as nutrient sink due to the progressive accumulation of litter. Nevertheless, in our case there is another factor to consider because the riparian area was composed mainly of planetree, which had important effects on leaf litter dynamics. Planetree leaves are easily scoured by the wind blow out of the riparian zone due to their shape and low permeability, and this effect is amplified by the periodic floods that are common in Mediterranean ecosystems (streamflow approximately three orders of magnitude greater than baseflow) (Casas and Gessner, 1999; Sabater et al., 2001). Therefore, in Fuirosos the forest floor acts as a sink of nitrogen, which is highly susceptible to scouring during windy and/or flood periods.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 SITE DESCRIPTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
In Mediterranean regions, where soil processes occur in pulses after rains, and floods are common, the type of riparian trees has a major importance in the ability of those systems to retain nitrogen either in vegetative structures or in the soil, or if, as occurs in the present study, the major part of the nitrogen returns in a form that is difficult to decompose and is highly susceptible to scouring to the streambed. The accumulation of nitrogen in tough litter in riparian corridors forested with planetree may provoke a gradual impoverishment of soils, which may have consequences for the structure of the riparian communities and their richness and diversity. The effect of afforestation with planetree may be attenuated if riparian corridors of original vegetation are kept in plantation forestry, as Abelho and Graça (1996) have also suggested for blue gum (Eucalyptus globulus Labill.).


   ACKNOWLEDGMENTS
 
The authors thank Antoni Bombí (Diputació de Barcelona, Servei Parcs Naturals) and Joan Cerrato-Gallego for their support in the field. This study was supported by funds provided by the European Community (NICOLAS Project ENV4-CT97-0395).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SITE DESCRIPTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 





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 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 (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bernal, S.
Right arrow Articles by Sabater, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bernal, S.
Right arrow Articles by Sabater, S.
Agricola
Right arrow Articles by Bernal, S.
Right arrow Articles by Sabater, S.
Related Collections
Right arrow Ecosystem Management
Right arrow Nitrogen
Right arrow Plant and Soil Interactions
Right arrow Nutrient Cycling

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
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome