Nitrogen Sediment Fluxes in an Upwelling System off Central Chile (concepción Bay and Adjacent Shelf) during the 1997-1998 El Niño

Two stations representing coastal and shelf environments (bay of Concepción and adjacent shelf, respectively) were sampled in order to assess benthic nitrogen fluxes. During the study period anomalously high oxygen bottom waters conditions were observed, coinciding with the 1997-1998 El Niño event. From both sites, intact sediment cores were collected for incubation experiments to directly estimate total ammonium fluxes. Pore-water chemistry was also assessed to estimate diffusive fluxes of nitrogen compounds (NH 4 + , NO 3-and NO 2-). The NH 4 + flux from incubation experiments ranged from 0.55 to 2.18 mmol m-2 day-1 in the inner bay and from 1.84 to 2.14 mmol m-2 day-1 on the shelf. NO 3-diffusive fluxes to the sediments were observed in the bay in March (-17 x 10-2 mmol m-2 day-1) and on the shelf in November (-2.7 x 10-2 mmol m-2 day-1), associated with the lowest NH 4 + diffusive fluxes and coincident with a stronger gradient of Eh in the first centimeters of the sediments. Besides, total NH 4 + fluxes lower than in normal periods were observed, suggesting the influence of less reducing conditions, consequence in turn of the lower fluxes of particulated organic material and higher dissolved oxygen in the bottom waters. These conditions were probably imposed by the El Niño event. On the other hand, and relative to previous reports, an increase of irrigation coefficients (DI) was estimated. Thus, the flux extension of other nitrogen compounds could have acquired relevance in relation to NH 4 + fluxes during the study period.


INTRODUCTION
The southeastern Pacific is one of the most productive systems in the world oceans.The continental shelf off central Chile at ~ 36° S extends about 40 km offshore and is one of the widest and most productive shelf areas along the Chilean coast (Scholl et al. 1970, Bernal et al. 1989, Dellarosa 1998).The hydrography of the area is dominated by seasonal and episodic coastal upwelling in response to wind patterns associated with the Pacific anticyclonic gyre (Ahumada et al. 1983).The combined effect of these winds and the Coriolis force gives rise to the upwelling of ESSW (Equatorial Subsurface Water) characterized by high salinity, low oxygen and high nutrient content (Ahumada & Chuecas 1979).The upwelled waters promote the phytoplanktonic growth and hence primary productivity.A maximum Chl-a concentration of about 50 mg m -2 (Ahumada et al. 1983, 1984, Ahumada 1989) and primary production values of 5.8 to 9.6 g C m -2 day -1 (Fossing et al. 1995, Dellarosa 1998) have been measured in this area.A large fraction (ca.25-40 %) of the primary production settles to the bottom (Bernal et al. 1989), where the major fraction of it is remineralized and its products, such as ammonium, released to the water column (Farías et al. 1996).
The warm phase of El Niño Southern Oscillation (ENSO) cycle in the tropical Pacific, i.e., El Niño, has been associated with climatic and economic impacts around the globe.El Niño has been described at several year intervals in the Peruvian-Chilean coasts (Enfield 1989, McPhaden 1999).This event is characterized by the presence of oxygen-rich and nutrient-poor waters that prevent high primary production during the upwelling periods.Off Concepción, during the 1997-1998 El Niño, warm waters and anomalous local winds (not favorable for upwelling) were observed, as well as high oxygen concentrations in the bottom waters (Gutiérrez 2000).Moreover, a substantially lower primary production, about 0.18-0.16g C m -2 day -1 (Ulloa personal communication) was measured.Other studies also indicate decreased quantity and quality of organic matter (Neira et al. 2001), compared to "normal" non-El Niño period.It is expected that these anomalous conditions could have promoted a different response of the biogeochemical process involved in the organic matter mineralization.These new conditions also affected the distribution and composition of macro -and meiofauna and, therefore, processes such as bioturbation (Gutiérrez et al. 2000, Neira et al. 2001), which is closely related to the magnitude of benthic fluxes.Thus, chemical and biological conditions during El Niño period could change the nutrient fluxes across the water-sediment interface.
The main objective of this study was thus to evaluate benthic nitrogen fluxes across the sediment-water interface in Concepción Bay (central Chile) during the 1997-1998 El Niño event.The results are also discussed in relation with existing data for the study area gathered during non-El Niño years.

Study site
The hydrography of the study area is closely linked to the adjacent oceanic system with regard to the interannual, seasonal and short time fluctuations, and to the local wind regime.Predominant southwest winds develop an upwelling center between 35° and 38° S (Brandhorst 1971, Ahumada & Chuecas 1979, Arcos & Salamanca 1984).On the seasonal time-scale, southwest winds usually blow for a week and are followed by a period of calm and northerly winds.Some authors have identified these periods as active and relaxed upwelling (Arcos & Wilson 1984, Peterson et al. 1988, Sobarzo et al. 1997).This pattern causes strong temporal variability in the water column hydrography.The upwelled waters (ESSW, Equatorial Subsurface Water) located between 100-400 m (Brandhorst 1971), can be found over the continental shelf and sometimes even at the inner Concepción Bay during active upwelling, causing subtoxic conditions over the whole area (Ahumada et al. 1983, Roa et al. 1995).

Sampling
Two study stations were selected, representing coastal (Station 4, 36°38' S, 73°02' W; 27 m deep) and shelf (Station 18, 36°30' S, 73°07' W; 88 m deep) environments.These were sampled in November 1997 and March 1998 (Fig. 1), where duplicate sediment cores (70 cm length, 10 cm inner diameter) were collected with a mini-multicorer (MUC).Only selected, undisturbed cores with clear overlying water were used.The cores were kept refrigerated (ca.3-4 °C), protected from light and transported within six hours as a maximum to the Marine Biological Station of Universidad de Concepción.Water samples were taken at standard depths using Niskin bottles and dissolved oxygen (DO) was measured in three replicates by the modified Winkler method (Williams & Jerkinson 1982) using a DOSIMAT for titration and a photoelectric cell for end point detection.

Pore water sampling and analysis
Sediment profiles for redox potential (Eh) were measured at intervals of 1 cm down to 20 cm using a platinum standard combination electrode with a calomel internal reference (Mettler Toledo-InLab).Duplicated cores from the same sampling location were analyzed for NH 4 + , NO 3 -and NO 2 -pore water concentration.Pore water samples were extracted by squeezing sediment sections of 1 cm under a N 2 atmosphere through a 0.45 µm mixed esters cellulose membranes (Gelman).The water thus obtained was immediately filtered again through GF/F filters (Whatman), stored in acid washed plastic vials and kept frozen (-20 °C) until later analysis.Nitrogen compounds (NO 3 -, NO 2 -and NH 4 + ) were analyzed using a Bran and Luebbe TRAACS 800 autoanalyzer.
Porosity and water content were determined at 1 cm intervals by weight loss after drying (70 °C) a known volume of sediment and assuming a sediment and water density of 2.65 g cm -3 and 1.025 g cm -3 , respectively (Buchanan 1984).
Pore water nutrient diffusion was estimated from pore-water concentration gradient between the top layer and 5 or 7 cm deep, according to Fick's first law (Berner 1980).A diffusion coefficient, corrected for temperature and tortuosity was used (Li 1974, Boudreau 1997).

Benthic flux experiments
Two 30 cm long, 8 cm inner diameter, undisturbed cores sub-sampled from original MUC cores were used for ammonium flux experiments across the water-sediment interface for each station.These cores contained approximately 15 cm of sediment covered by 15 cm of water (total volume 600 mL).The overlying water was carefully replaced with bottom water before incubation.Cores were wrapped with aluminum foil to exclude light, and incubated for 6 h at the in situ bottomwater temperature (~ 11 °C) after collection.Each core was capped and equipped with dissolved oxygen, temperature and pH sensors, and a magnetic stirrer for mixing during 6-8 h incubations.Initial ammonium samples were taken from the in situ bottom water and successive samples were obtained at different time intervals (every 10 min the first hour and from 10 to 20 min the next 5 h).Concentrations were measured according to the modified Solorzano indophenol method (Grasshoff 1983).An analytical error of 3 % was estimated using ammonium chloride standards.
Sediment-water fluxes in the incubated cores were estimated directly based on the best fit of the obtained time course curves, according with the best variance explained.A two way Analysis of Variance (ANOVA) was used to assess differences in ammonium total fluxes between sampling periods and stations.When significant differences were found, contrasts between groups were performed using the Tukey HSD a posteriori test (Zar 1974).The STATISTICA software was used for these purposes.

Water column and sediment environmental settings
The time course of the temperature and dissolved oxygen (DO) concentration in the area between May 1997 and March 2000 are shown in Fig. 2. A slight warming, mainly at the surface and anomalously high bottom water oxygen concentrations between November 1997 and May 1998 were evident.One year later, between November 1998 and March 1999, low oxygen concentrations (<10 mM) and colder bottom water (around 11 °C) conditions, considered typical for the area, returned.These anomalous conditions between May 1997 and May 1998 were attributed to the El Niño 1997-1998 event.
The redox potential (Eh) and the bottom water DO values during this study are shown in Table 1.For practical purposes and to avoid large tables, the sediment column was divided in two layers: surface (0-3 cm) and bottom (> 3 cm), coincident with the redox boundary (change between positive and negative values) in two cases.Sediment layers with Eh values lower than -100 mV are indicated at the footnote.Differences between stations and periods were observed.Negative values of Eh were found in the bay sediments during November, while bottom water depicted oxygen concentrations of 98.1 mM, which are unusually high for this period.During March, less reduced surface sediments (0-3 cm) were observed within the bay as well as lower bottom water DO concentrations (16.9 mM).At the shelf site, sediments presented positive Eh surface values during both samplings.Bottom water DO concentrations varied from 25.4 to 16.4 mM, November 1997 to March 1998, respectively.

Pore water concentrations
The vertical distribution of solutes in pore water is shown in Fig. 3 (November 1997) and Fig. 4 (March 1998).They represent the mean value estimated from two cores obtained simultaneously (mean ± SD).The pore water NH 4 + concentrations show the typical vertical gradient, increasing in depth in both sampling periods and stations.The NO 3 -and NO 2 - concentrations decreased in depth from the sediment-water interface or with the presence of subsurface maxima at 2-3 cm depth.The NO 3 -concentration profiles in the bay sediments (Station 4) in November showed a different pattern; concentrations increased in depth but with a notorious decrease between 4 and 8 cm.From this depth to the bottom of the core, concentrations increased again and reached a maximum value of 7 mM.The NO 3 - concentrations at surface sediments of the bay (5 mM) were lower than in the shelf (20 mM).However the NH 4 + surface concentrations varied between 59 and 60 mM in both areas.In March, NH 4 + surface concentration was higher at the bay (Station 4) (~ 120 mM) with NO 3 - concentration also showing a conspicuous increment at the same time in this station (~ 89 mM).NO 2 -concentrations were similar in both periods but being lower at the bay (Station 4).

Diffusive fluxes
The estimated diffusive fluxes of nitrogen compounds are shown in Table 2.These values correspond to the mean values ± SD, determined from two cores obtained simultaneously.The pore water pool of NH 4 + at the bay was higher than at the shelf, in contrast the NO 2 -pool was higher at the shelf, both not showing differences between sampling periods.Differences in NO 3 -pool were observed only at the bay, with highest values occurring during March (Table 2).
In general, the relatively higher values for NH 4 + diffusive fluxes suggest that this compound dominates the exchange of nitrogen between the sediment and the water column.However, the NO 3 -flux on the shelf (Station 18) during March 1998 is in the same order of magnitude as the NH 4 + flux, indicating that the production of NO 3 -in the sediments acquires relevance.Also, the negative fluxes of NO 3 -at the bay and shelf during March and November respectively, were associated with the lowest NH 4 + diffusive fluxes coincident with the  Gutiérrez (2000), Chl-a content at 0-1cm, H 2 S inventory at 0-15 cm depth e. Gallardo et al. (unpublished data).Total mass of bacteria expressed in g m -2 present at 0-15 cm, mean ± SD from 3 cores Concentración de compuestos nitrogenados en el agua intersticial en noviembre de 1997.La línea punteada señala la extensión de la capa oxidada.Las barras de error son la desviación estándar del análisis de dos replicas y la línea vertical el espesor de la sección.
stronger gradient of Eh in the first centimetres of sediments, which is more evident in the bay sediments (+ 41 to -99 mV) (Table 1).

Benthic ammonium fluxes
The time course curves are shown in Fig. 5.They were corrected for the initial ammonium concentration values (time zero) in the experiments.These initial values were ~ 27 and 16 mM in November and in March they were 12 and 16 mM, in the bay and shelf sediment experiments, respectively.The concentration showed high variability during the first hour, therefore a good fit was not achieved for some curves (variance between 56 and 91 %) (Table 3).The slope of the curves suggests that the fluxes increased in time, with seasonal differences in the bay sediments, but not in the shelf.The NH 4 + fluxes in the shelf station (Station 18) during both periods were not significantly different from fluxes estimated in the bay (Station 4) during March (ANOVA, F 3,4 = 1.95,P < 0.1), probably due to the fact that bottom water showed low variability in oxygen content (Table 1).In contrast, the flux of the bay sediments in November was the lowest and significantly different (0.55 ± 0.11 mmol m -2 day -1 , Tukey HSD test, P < 0.1) concurrent with a maximum concentration of bottom-water dissolved oxygen (Table 1).

DISCUSSION
A deepening of the oxidized layer was observed in both sites, consistent with the fact that anoxic conditions (in contrast with what happens during normal years; Ahumada & Chuecas 1979) did not develop by the end of summer (Table 1).Besides, more aeration by benthic faunal activity was produced (Gutiérrez et al. 2000), which Concentración de compuestos nitrogenados en el agua intersticial en marzo de 1998.La línea punteada señala la extensión de la capa oxidada.Las barras de error son la desviación estándar del análisis de dos replicas y la línea vertical el espesor de la sección.
29 ± 6.3x10 -4 0.03 ± 0.01 0.09 ± 0.01 11 ± 1.5x10 -2 1.7 ± 0.4 0.9 ± 0.2 NO 2 - -4.2 ± 1.7x10 -4 8.9 ± 4.4x10 -3 0.03 ± 0.01 0.13 ± 0.06 Values ± SD from two cores a Estimated at 0-15 cm depth could have affected the pore-water chemistry of the bay sediments.In fact, the NH 4 + concentrations in the bay (~50-90 mM, surface sediments) are lower than in previous reports (~200-600 mM; Farías & Salamanca 1990, Farías et al. 1995), in consequence lower pore water pool of NH 4 + are observed than those reported during no El Niño conditions (between 20 to 142 mmol m -2 , Graco et al. 2001) (Table 2).Nitrification should be favored in oxic conditions.NO 2 -profiles showed relatively low concentrations with maximum values around 2 cm deep and just above the depth where NO 3 - reaches the maximum.This constitutes the typical profile that shows NO 2 -as an intermediate product in the nitrification process (Atlas & Bartha 1998).However, the influence of dissolved oxygen in these processes is not clear.The NO 3 -pore water pools were higher in March at the bay sediments, where the highest NO 3 -concentrations and a deeper oxidized layer were observed but not coincident with the maximum DO at the bottom waters.This indicates a response time since water column changed environmental conditions appear and the effect in the sediments develop.The bay sediments seem to be more affected than the shelf sediments due to they are normally under highly reducing conditions (Eh < -100mV).
Positive correlations between nitrogen species (NO 3 -and NH 4 + ) and bottom-water oxygen have been reported in shallow waters (Cowan et al. 1996, Graco et al. 2001), reflecting the sensitivity of nitrification and denitrification processes relative to DO concentrations.It is possible that NO 3 - formation could occur more favourably in superficial regions of the sediments where the overlying water is more oxygenated supplying oxidant (O 2 or Mn, Hulth et al. 1999).Therefore the occurrence of increased bottomwater oxygen (> 45.4 mM) could promote nitrification in the upper sediment layers, especially at the bay sediment.

Diffusive fluxes
The NO 3 -negative fluxes and related lower NH 4 + fluxes in the bay and shelf during March and November, respectively (Table 2), could be indicating a response to a nitrification and a rapid denitrification process (Hulth et al. 1999), favored by sulphide inventories in the sediments lower than in normal periods (Table 1; 192.4 mmol m -2 during non-El Niño, Gutiérrez 2000), since the high sulphide concentrations inhibit both processes (nitrification-denitrification, Joye & Hollibaugh 1995).Also, the relatively lower Flujos de amonio determinados en los testigos de incubación 1 y 2 en la estación 4 (1A and 2A) y 18 (1B and 2B) en ambos períodos de muestreo.
NO 3 -fluxes could have been due to denitrification as a dominating NO 3 --consuming process in the sediment (Jensen et al. 1990, Hulth et al. 1999, Usui et al. 2001).However, though the N 2 O produced in this process would have been an important clue to clarify this point, it was not measured in this study.At the shelf the negative flux could also result from an uptake of the overlying water combined with NO 3 -production from within the sediments with a positive redox potential (Jenkins & Kemp 1984); actually the NO 3 -concentration in the superficial sediments is similar to values found in the shelf bottom water (Gutiérrez 2000) (Table 1).
The differences in NO 3 -and NH 4 + fluxes between stations and sampling periods could be related also with the organic matter availability.In March 1998 a weak reactivation of upwelling was observed (Gutiérrez 2000) probably due to the development of an algae bloom.Actually, higher Chl-a concentrations in superficial sediments of the bay (Station 4) were observed compared to the previous period (November 1997, Gutiérrez 2000) (Table 1).Probably the shallowness of the bay could decrease the oxygen penetration reducing nitrification as observed after a bloom in shallow anoxic sediments (Jenkins & Kemp 1984, Jensen et al. 1990).On the shelf sediments, the mineralization of sedimented organic matter is conducted in a permanent more oxygenated environment.Thus, the organic material availability associated with the redox potential, which determines the oxidative pathways of organic matter mineralization (Atlas & Bartha 1998, Usui et al. 2001), is an important factor controlling the nitrogen species abundance in both areas.Besides, the differences in the diffusive fluxes are responding to the differences observed in the environmental conditions in both sampling areas (bay and shelf) emphasized by the shallowness of the bay (Table 1 and 2).
Temporal differences in the bay (Station 4) attributable to the activity of the benthic bacterium Beggiatoa sp.mats are discarded.During sampling period, and also in response to El Niño-changed conditions, lower bacterial biomasses were observed.Reported values for the summer season range from 29-35 g m -2 for Beggiatoa sp. to ~202-774 g m -2 for Thioploca sp.(Fossing et al. 1995, Graco et al. 2001).These bacteria can accumulate high intracellular concentrations of NO 3 -(300-500 mM) and elemental sulfur (Fossing et al. 1995, Jørgensen & Gallardo 1999), being responsible for the NO 3 -fluxes into the sediments (consumption) (Graco et al. 2001).In this study, the highest negative NO 3 diffusive fluxes were estimated in March 1998 (-17 x 10 -2 mmol m -2 day -1 ) while a positive flux was found during November 1997.Beggiatoa sp. was present only in March 1998 (absent during November 1997) (Table 1) suggesting an uptake of NO 3 -related with the activity of this bacterium.However, at the shelf, the NO 3 -diffusive fluxes where negative (uptake) in November and positive in March (-2.7 and 11 x 10 -2 mmol m -2 day -1 , respectively) besides Thioploca sp. was present in both sampling periods, though in low biomasses (Table 1).Thus, its effect over the fluxes could not be observed clearly.
The diffusive fluxes estimated through solute concentration gradients do not consider the bioturbation effects (advective flux).Though this results in lower values than the total fluxes measured through sediment core incubations, the diffusive fluxes provide a general picture of the biogeochemical processes occurring in the sediments.

Benthic ammonium flux at the sediment-water interface
Previous reports on ammonium fluxes within the bay indicated higher values, between 5 and 9 mmol m -2 day -1 (Farías et al. 1996, Thamdrup & Canfield 1996), when the normal "anoxic" conditions were present.During the sampling period a lower amount of organic matter was settled with oxic conditions prevailing, evidenced this by lower Chl-a concentrations in surface sediments (Table 1) compared to those reported for other periods (~165 and 1,120 mg g -1 in November 1998and March 1999, respectively, Gutiérrez 2000, Graco et al. 2001).This produces changes in the biogeochemical processes in the sediments affecting mainly the NH 4 + concentrations since these are highly sensitive to changes of organic matter in the sediments (Blackburn & Henriksen 1983, Jensen et al. 1990).
Although the ammonium fluxes from sediment incubation experiments were low, they indicate that the sediment continue representing a source of ammonium to the overlying water during El Niño.The relative increase in the flux of other nitrogen species in comparison to "normal" periods is not discarded.If we consider that the water column production is ~ 2-9 g C m -2 day -1 over the shelf, and ~ 3.5 g C m -2 day -1 in the bay in normal years (annual basis) (Pantoja et al. 1987, Fossing et al. 1995, Dellarosa 1998), and assuming (1) 25-40 % loss in the sediments (Bernal et al. 1989, Farías 1998) and (2) ~ 80-90 % of N is regenerated from Concepción Bay sediments during the spring-summer period (Farías et al. 1995), then a nitrogen flux between 4 to 32 mmol N m -2 day -1 would have been expected in both areas (assuming a C:N ratio equivalent to Redfield).In March 1998, primary production of about 0.17-0.18g C m -2 day -1 was determined in the shelf area (Ulloa personal communication).Using the same criteria described above, the nitrogen flux should have been 0.4-0.7 mmol N m -2 day -1 , in the lower range to that estimated from the ammonium incubation experiments (Table 3).However this approximation does not reflect the processes occurring within the sediments.The mineralization of organic material could have been stimulated during the study period.Normally, these sediments show relatively high C:N ratios compared to those found between 1997 and the first half of 1998 (Gutiérrez et al. 2000, Neira et al. 2001), suggesting the presence of lower amounts of refractory organic carbon in the sediments during this study.This implies that when anomalous oxygenation occurs, a mineralization of previously deposited organic mater is developed.The nitrogen flux directions as NO 3 - could change or increase, representing an influx during normal periods in the area (Graco et al. 2001).Direct measurements of benthic NO 3 - fluxes were not performed in this study, but they could have represented a flux from the sediments, as suggested by the estimated diffusive fluxes.
Thus, during the study period the nutrient fluxes did not reflect only the degradation of recent settled organic material.Ventilation and feeding activity stimulate the microbial processes increasing the nutrient release from sediments (Kristensen et al. 1991, Christensen et al. 2000).In the bay (Station 4, lowest flux), significantly higher mixing rates were observed in November 1997 in agreement with higher bioturbation potential observed.In the shelf (Station 18) instead, the same bioturbating species dominated during the whole sampling period (Gutiérrez et al. 2000) with a bioturbation potential lower than in the bay.Thus, the fluxes found in this study are also a result from this variability, which in turn is related to the oxygen content and to the organic matter availability.The denitrification is another important process that could be increased during high bioturbation activity (Kristensen et al. 1991) in the bay, which supplies oxidants to the underlying anoxic regions.This process was probably higher during El Niño, but it could not be estimated in this study.

Approach of the 1997-1998 El Niño effects on the total fluxes
The incubation chamber fluxes represent a reasonable approach to the total flux from the sediments (Table 3).They result from several processes acting at the same time and in two levels, i.e., molecular and advective.The molecular flux is proportional to the vertical gradient of dissolved solute concentrations described by the first Fick's law (Li 1974, Berner 1980, Boudreau 1997), depending on the diffusion coefficient (D S ).The advective flux depends on the gradient determined by particle reworking and fluid motion due to bioturbation, quantified through a biodiffusive coefficient (D B ) and irrigation coefficient (D I ).Thus, the total flux (J t ) is given by: J t = φ (D S + D B + D I ) ∂C/∂z Where, φ is porosity and ∂C/∂z is the concentration gradient.The irrigation flux can be estimated by difference using the above relation.D B coefficients estimated for the same sampling period are between 1.1-5.6 x 10 -6 cm 2 s -1 in bay sediments and 1.5 x 10 -6 cm 2 s -1 in the shelf (Gutiérrez 2000).Thus, the calculated NH 4 + advective flux due to irrigation varied between 0.31 and 2.13 mmol m -2 day -1 in bay sediments and between 1.78 and 1.90 mmol m -2 day -1 in shelf sediments.This compared to the total fluxes measured (Table 3), suggests that the irrigation is the main process regulating the exchange of NH 4 + in the sediment-water interface.Similar results were found in the bay during non-El Niño conditions (Farías et al. 1995).However, the irrigation coefficients (D I ) estimated in this study, ~0.2-3 x 10 -4 cm -2 s -1 in the bay sediments and 1-9 x 10 -4 cm -2 s -1 in the shelf sediments, are one order of magnitude higher than those estimated previously in the bay (1.5-3.4 x 10 -5 cm 2 s -1 ; Farías et al. 1995), suggesting that the fluxes are more influenced by the benthic animal activities in this period.These results are in agreement with those found by others, where an increased bioturbating activity and an enhanced vertical penetration of infauna during El Niño is determined, specially in shallow zones of the bay and at the greatest depth over the shelf (Gutiérrez et al. 2000, Neira et al. 2001).This change was associated with oxygen levels higher than in normal years and with reduced fluxes of bulk and reactive organic carbon to the sediments.Thus, the increase of the organic matter remineralization is facilitated by the oxygenation of the sediments through the activities of benthic animals.Also, in the sediments under anoxia for large periods, the exposition to oxygen by resuspension, diffusion or bioturbation, promotes mainly the oxidation of refractory material, enhancing the extent of organic carbon degradation (Hulthe et al. 1998).
The results of this study suggest that the lower exchange of NH 4 + in the sediment-water interface, relative to previous reports, is associated to the anomalous higher DO content in the bottom water, consequence in turn of the 1997-1998 El Niño event.This could also support that nitrification process in the top centimeters of the sediment occurred, suggested by the high pore-water nitrate concentrations measured.A potential increase in other nitrogen compound fluxes in relation to the low NH 4 + fluxes, in spite of the lower primary productivity, is expected due to the enhanced organic matter mineralization within the sediments.

Fig. 3 :
Fig. 3: Concentration of nitrogen compounds in pore waters in November 1997.Doted line shows the extent of oxidized layer.Error bars (horizontal) are the standard deviation of analyses from two replicates and vertical lines indicate section thickness.

Fig. 4 :
Fig. 4: Concentration of nitrogen compounds in pore waters in March 1998.Doted line shows the extent of oxidized layer.Error bars (horizontal) are the standard deviation of analyses from two replicates and vertical lines indicate section thickness.

TABLE 1
Some relevant parameters describing the environmental conditions in the sediments of ConcepciónBay and adjacent shelf during the sampling period Algunos parámetros relevantes que describen las condiciones ambientales en los sedimentos de la Bahía Concepción y plataforma adyacente durante el período de muestreo

TABLE 2
Diffusive fluxes and size of nitrogen compounds in pore waters estimated on sediments of Concepción Bay and adjacent shelf.Values are means ± SD from two cores Flujos difusivos e inventarios de compuestos nitrogenados estimados en los sedimentos de la Bahía Concepción y plataforma adyacente.Los valores corresponden a medias ± DE

TABLE 3
Ammonium fluxes from incubation cores in the study area.Values are means ± SE of curve fit Flujos de amonio estimados por incubación de sedimento en el área de estudio.Los valores corresponden a medias ± EE de la curva ajustada Values ± standard error of curve fit mean ± SD