Volume 18, numéro 4, 2005

FR :

Actuellement, en considérant simultanément les éléments constitutifs du risque, soit l'aléa et la vulnérabilité, aucune des méthodes existantes dites de cartographie des risques d'inondation ne permet d'établir de façon précise et quantifiable en tous points du territoire les risques d'inondation. La méthode de cartographie présentée permet de combler ce besoin en répondant aux critères suivants : facilité d'utilisation, de consultation et d'application, résultats distribués spatialement, simplicité de mise à jour, applicabilité à divers types de résidences.

La méthode présentée utilise une formulation unitaire du risque basée sur les taux d'endommagement distribués et reliés à diverses périodes de retour de crues à l'eau libre. Ceux-ci sont d'abord calculés à partir des hauteurs de submersion qu'on déduit de la topographie, des niveaux d'eau pour des périodes de retour représentatives et du mode d'implantation des résidences (présence de sous-sol, hauteur moyenne du rez-de-chaussée). Ensuite, le risque unitaire est obtenu par intégration du produit du taux d'endommagement croissant par son incrément de probabilité au dépassement. Le résultat est une carte représentant le risque en % de dommage direct moyen annuel. Une étude pilote sur un tronçon de la rivière Montmorency (Québec, Canada) a montré que les cartes sont expressives, flexibles et peuvent recevoir tous les traitements additionnels permis par un SIG tel que le logiciel MODELEUR/HYDROSIM développé à l'INRS-ETE, l'outil utilisé pour cette recherche. Enfin, l'interprétation sur la Montmorency des cartes d'inondation en vigueur actuellement au Canada (les limites de crue de 20/100 ans) soulève des interrogations sur le niveau de risque actuellement accepté dans la réglementation, surtout quand on le compare aux taux de taxation municipale.

EN :

Public managers of flood risks need simple and precise tools to deal with this problem and to minimize its consequences, especially for land planning and management. Several methods exist that produce flood risk maps and help to restrict building residences in flood plains. For example, the current method in Canada is based on the delineation in flood plains of two regions corresponding to floods of 20- and 100-year return periods (CONVENTION CANADA/QUÉBEC, 1994), mostly applied to ice-free flooding conditions. The method applied by the Federal Emergency Management Agency FEMA (2004) is also based on the statistical structure of the floods in different contexts, with a goal mostly oriented towards the determination of insurance rates. In France, the INONDABILITÉ method (GILARD and GENDREAU, 1998) seeks to match the present probability of flooding to a reduced one that the stakeholders would be willing to accept.

However, considering that the commonly accepted definition of risk includes both the probability of flooding and its consequences (costs of damages), very few, if any of the present methods can strictly be considered as risk-mapping methods. The method presented hereafter addresses this gap by representing the mean annual rate of direct damage (unit value) for different residential building modes, taking into account the flood probability structure and the spatial distribution of the submersion height, which takes into account the topography of the flood plain and the water stage distribution, the residential settlement mode (basement or not) and the first floor elevation of the building. The method seeks to meet important criteria related to efficient land planning and management, including: ease of utilisation, consultation and application for managers; spatially distributed results usable in current geographical information systems (GIS maps); availability anywhere in the area under study; ease of updating; and adaptability for a wide range of residence types.

The proposed method is based on a unit treatment of the risk variable that corresponds to a rate of damage, instead of an absolute value expressed in monetary units. Direct damages to the building are considered, excluding damages to furniture and other personal belongs. Damage rates are first computed as a function of the main explanatory variable represented by the field of submersion depths. This variable, which is obtained from the 2D subtraction of the terrain topography from the water stage for each reference flood event, is defined by its probability of occurrence. The mean annual rate of damage (unit risk) is obtained by integrating the field of damage rate with respect to the annual probability structure of the available flood events. The result is a series of maps corresponding to representative modes of residential settlement.

The damage rate was computed with a set of empirical functional relationships developed for the Saguenay region (Québec, Canada) after the flood of 1996. These curves were presented in LECLERC et al. (2003); four different curves form the set that represents residences with or without a basement, with a value below or above $CAD 50,000, which is roughly correlated with the type of occupation (i.e., secondary or main residence). While it cannot be assumed that theses curves are generic with respect to the general situation in Canada, or more specifically, in the province of Québec, the method itself can still be applied by making use of alternate sets of submersion rates of damage curves developed for other specific scenarios. Moreover, as four different functional relationships were used to represent the different residential settlement modes, four different maps have to be drawn to represent the vulnerability of the residential sector depending of the type of settlement. Consequently, as the maps are designed to represent a homogeneous mode of settlement, they represent potential future development in a given region better than the current situation. They can also be used to evaluate public policies regarding urban development and building restrictions in the flood plains.

A pilot study was conducted on a reach of the Montmorency River (Québec, Canada; BLIN, 2002). It was possible to verify the compliance of the method to the proposed utilisation criteria. The method proved to be simple to use, adaptive and compatible with GIS modeling environments, such as MODELEUR (SECRETAN at al, 1999), a 2D finite elements modeling system designed for a fluvial environment. Water stages were computed with a 2D hydrodynamic simulator (HYDROSIM; HENICHE et al., 1999a) to deal with the river reach complexity (a breaded reach with back waters). Due to the availability of 2D results, a 2D graphic representation of the information layers can therefore be configured, taking into account the specific needs of the interveners. In contexts where one dimensional water stage profiles are computed (e.g., HEC-RAS by USACE, 1990; DAMBRK by FREAD, 1984), an extended 2D representation of these data needs to be developed in the lateral flood plains in order to achieve a 2D distributed submersion field.

Among the interesting results, it was possible to compare the risk level for given modes of settlements (defined by the presence/absence of a basement and the elevation of the first floor with respect to the land topography) with current practices, based only on the delineation of the limits of the flood zones corresponding to 20/100 year return periods. We conclude that, at least in the particular case under study, the distributed annual rate of damage seems relatively large with respect to other financial indicators for residences such as urban taxation rates.

FR :

Cet article présente les conclusions d'une recherche visant l'amélioration de l'annonce des crues, et son application à la rivière Thoré, dans le contexte du système d'alerte français. On y exploite les informations météorologiques contenues couramment dans les bulletins d'alerte aux précipitations [BAP] émis par Météo-France, dans le but d'aider les prévisionnistes du Service d'annonce de crues [SAC] à anticiper l'atteinte de la cote d'alerte sur une rivière. Le travail présenté fait partie d'une approche visant à munir les SAC d'outils prévisionnels fonctionnant en temps réel et aptes à prévenir d'une mise en alerte probable. L'approche préconisée conduit à une utilisation directe des informations contenues dans les BAP reçus des services de météorologie dans le processus de surveillance des crues. C'est au moyen de courbes d'intensité-durée-temps d'alerte [IDTA], préalablement établies pour des prévisions de pluies uniformément réparties, et de courbes d'intensité-superficie-temps d'alerte [ISTA] pour les prévisions relatives à des cellules orageuses localisées, que l'approche proposée est développée.

EN :

This work was designed to contribute to the improvement of flood forecasting, in the context of the French alert system. We propose that the meteorological information contained in the French weather forecast bulletin (Bulletins d'alerte aux précipitations; BAP), produced by Météo-France (French meteorological organization), should be utilized in order to aid the forecasters of the French flood forecasting agencies (Services d'annonce de crues; SAC) to anticipate the timing of an alert associated with an increase in the water level of a river. The goal was to develop an approach to provide the SAC with a real-time operational forecasting tool in order to improve the evaluation of a probable Flood Alert decision. This approach integrates the information contained in the BAP received from Météo-France into the existing flood control process with the use of Duration-Intensity-Warning Time (durée-intensité-temps d'alerte, IDTA) curves for uniform rainfall forecasting, and Intensity-Area-Warning Time curves (intensité-superficie-temps d'alerte, ISTA) for localized storm cells.

The rainfall parameters considered were the intensity (I, mm/h), the duration (D, h), and the area of the watershed affected by the rainfall (S, km²). These parameters are related by the equation V=I x D x S, where V is the volume of rain (hm3). The parameter directly related to the Flood Alert decision is the warning time (T_alerte), measured in hours. It is defined as the time from the beginning of the rainfall to the time when the flow at the watershed outlet reaches the alert flow (Q_alerte in m³ /s), regardless of the maximum discharge (Q_max). Although Qmax may be an important indicator of the magnitude of the upcoming event, the chief concern is the Flood Alert decision, and therefore, the time to alert parameter (T_alert) is of primary importance.

The proposed approach involves creating a graphical connection of a series of rainfall intensity values (I) as a function of a range of rainfall (D) duration values with time to alert (T_alert) curves, which represent the I-D couples. As a result, a SAC forecast agent that receives a BAP indicating the quantitative precipitation forecast in a precise region for a defined period will be able to evaluate the time after the start of the rainfall that the alert flow (Q_alert) will be reached, simply by referring to the IDTA and/or ISTA curves. If an alert is foreseen within a certain delay, the flood forecast agent can wait to receive improved forecasts before making the decision whether to start the flood alert procedures or not.

The construction of the IDTA and ISTA curves requires numerous simulations in order to cover a wide variety of intensity-duration and intensity-area of rainfall couples for which the alert flow (Q_alert) will be reached at the watershed outlet, and therefore the time corresponding to this discharge can be estimated. The simulations were performed through the use of a combination of a deterministic distributed parameter hydrological model and a hydraulic one-dimensional hydrograph transfer model. The neuronal models of the Generalized Regression Network (GRNN) type were also used. This allowed for the extraction of and/or interpolation between values in the database containing the parameters intensity, duration, area, and the hydrographs that resulted from the simulations done with the first two models. The interest in using the GRNN model is to cover a large range of values for all of the parameters considered, without having to simulate all cases, therefore reducing the potential computation time.

We developed this forecasting approach on the basis of a specific case related to the extreme flooding that occurred in southern France on November 1999. More precisely, our case study concerns the mountainous region in the upstream area of the Thoré watershed, in the Tarn Department. The simulation scenarios were 1) uniformly distributed rainfall on a watershed of 208 km² ; 2) storm cells of 9, 36, 64 and 144 km² located in the watershed center; and 3) a storm cell located in various zones of the watershed.

The main observation of the simulation results was that the Talert was constant for a rainfall of intensity I, as long as the duration was longer than the T_alert (i.e., provided it was still raining after Q_alert was attained at the outlet). On the other hand, if the rain stops before Q_alert is attained, T_alert is delayed. T_alert increases as a function of the duration of the rainfall, for a constant I. This is true for both uniform and localized rainfall.

The IDTA and ISTA curves were developed on the basis of several simplifying hypotheses and should be improved in order to increase their precision and flexibility. Therefore this approach can be amended by taking into account the following factors:

- infiltration (when the laws defining it are established);

- the initial conditions: since the results of the simulations for this study are valid for constant initial conditions of Qini=20 m³/s, it would be pertinent to include a correction factor to adjust the results (T_alert) for the real initial conditions such as the actual Qini and the actual soil humidity;

- the spatial variability of the storm cells; and

- the combination of uniformly distributed rainfall and localized storm cells.Nevertheless, we evaluated the use of the forecasting approach with the IDTA and ISTA curves referring to the November 1999 events. The contribution of these curves in the Flood Alert decision process was assessed with a fictitious scenario defined by the issued BAP related to this event. Understanding the simplifying hypotheses discussed above, we conclude that the flood alert on the Thoré River watershed could have been advanced up to seven hours and thirty minutes from the actual time it was issued. In a fast or flash flood event, this range of anticipation could have a considerable impact.

FR :

Dans ce travail est présentée la composition biochimique en acides aminés (AA) libres et combinés et en vitamines de Micractinium pusillum isolée à partir du lagunage naturel expérimental de Ouarzazate où cette microalgue prolifère pendant la période chaude. M. pusillum est produite en culture en vrac, sur milieu synthétique, dans des conditions de lumière et de température fixes. Cette étude révèle la richesse de M. pusillum en acides aminés combinés qui constituent environ 59% de sa matière sèche. Les acides aminés libres forment 3,4% de cette matière. Dix-sept acides aminés et quatre vitamines (D3, B2, B3, et B12) ont été déterminés chez cette microalgue.

EN :

The biotechnological exploitation of microalgae requires more investigations in order to improve our knowledge of their biochemistry and to look for new potentially interesting species. Microalgae are used in various areas. Among them, wastewater treatment in wastewater stabilisation ponds and high-rate ponds is widely prevalent. Beside their purifying ability, their abundance in these ecosystems makes them interesting as a potential source of valuable products.

Micractinium pusillum Fres, a green alga (Chlorophyceae) that grows in these ecosystems, was isolated from the experimental wastewater stabilisation pond of Ouarzazate city (south of Morocco), where it becomes important during the hot period. In the present work, we investigated the dissolved free and combined amino acids and the vitamins produced by this microalgal strain cultured on a mineral medium. The biomass production of M. pusillum was conducted in 2 l cotton-stoppered flasks under continuous light (60 µE m^-2 s^-1) provided by neon tubes and at a temperature of 24 ± 2°C. The culture medium was sterilised by autoclaving before inoculation. Free and combined amino acids were respectively analysed by HPLC separation of fluorescent o -phthaldialdehyde (OPA) derivatives. Prior to derivatization, the cells were broken and the cell content was digested with HCl. The analysis of vitamins was also realised by HPLC, with UV - visible detection methods, after extraction by appropriate solvents.

The results reveal a high content of combined amino acids of approximately 59% (dry matter). Free amino acids constitute 3.4% (dry matter). Seventeen amino acids have been identified in this microalga (Table 1): aspartic acid, glutamic acid, arginine, methionine, phenylalanine, alanine and lysine were the principal components. The analysis of spectra indicated the presence of six categories of amino acids (AA) (Table 2): aliphatic AA, hydroxy AA, sulphur-containing AA, basic AA, aromatic AA and the diacid AA. Aliphatic AA (glycine, valine, leucine, and isoleucine) largely dominated the total AA pool, with more than 27% of total AA. Diacids AA (aspartic and glutamic acids), at approximately 25%, came in the second rank. Dibasics AA (15%) were relatively more abundant than aromatics AA (11%) and sulfur-containing AA (8%). It should be noted, however, that among the sulphur-containing AA, only methionine was detected by the method used in this work. Hydroxy AA are much less represented with only 4% of the total AA.

Four vitamins were detected in M. pusillum (Table 3): vitamins D3, B2, B3, and B12. The content of vitamin D (6970 µg g^-1) was highest, followed by vitamin B3 (2820 µg g^-1) and vitamin B12 (1860 µg g^-1). Vitamin B2 constituted 406 µg g^-1. According to the reported values of vitamin contents in several other microalgal species, as well as some conventional foods (Table 5), it appears that M. pusillum species contains relatively important amounts.

FR :

Les communautés de rotifères (Brachionidae et Trichocercidae) et de cladocères peuplant les eaux du lac Municipal de Yaoundé situé en zone tropicale (Cameroun), ont été examinés en relation avec quelques variables physico-chimiques du milieu. Des 26 espèces et sous-espèces de rotifères (Brachionidae et Trichocercidae) et 15 espèces de cladocères inventoriées, environ 16 sont identifiées pour la première fois au Cameroun. Parmi les espèces quantitativement dominantes, les espèces fréquemment rencontrées sont Brachionus angularis angularis, B. falcatus falcatus, et B. calyciflorus chez les Brachionidae, Trichocerca elongata elongata, et T. bicristata bicristata chez les Trichocercidae, et Ceriodaphnia cornuta et Chydorus eurynotus chez les cladocères. Plus de 75% des espèces répertoriées sont des espèces périphytiques communément considérées comme littorales, dont le développement important a été observé dans la zone pélagique du lac qui, sans doute, offre de nombreuses ressources et constitue un refuge vis-à-vis notamment de la prédation. L'absence de corrélations entre les communautés zooplanctoniques étudiés et les quelques variables physico-chimiques suivies, nous a amenés à considérer que le développement de ces communautés est sous la dépendance d'autres facteurs. D'ailleurs, une analyse canonique de correspondance indique que les stations, profondeurs et mois de prélèvement expliqueraient respectivement 3%, 5% et 36% de la variance totale associée à l'abondance des espèces prises en compte. Ce qui nous a permis d'émettre l'hypothèse selon laquelle les principales conditions environnementales qui affectent le développement potentiel des espèces analysées et le déroulement de leurs cycles biologiques, seraient dépendantes des deux saisons (la saison des pluies et la saison sèche) caractérisant les climats tropicaux.

EN :

The rotifer and cladoceran communities of a small tropical eutrophic lake (the municipal lake of Yaoundé, Cameroon) were examined from November 1996 to December 1997, in relation to some physical and chemical variables. Using a Van Dorn bottle, water samples were collected weekly from 5 depths (0 m, 0.5 m, 1 m, 1.5 m and 2.5 m) at three stations (I, II and III) representing the upper, middle and lower parts of the lake, respectively. Water temperature was measured with a thermometer placed in the sampling bottle and pH was measured in the field with a portable pH-meter. Oxygen concentrations were determined according to the Winkler method. Water colour, suspended solids, and ammonium-nitrogen (NH₄ -N) were analyzed spectrophotometrically using standard methods. The transparency of the water column was determined with a Secchi disk. Zooplankton species were identified alive under a phase contrast microscope using several manuals, and rotifers and cladocerans were counted in a Dolfuss chamber

Throughout the study, temperature and pH values of the water remained around 25°C and 7, respectively. The Secchi disk transparency did not exceed 1 m and the dissolved oxygen concentration decreased with depth to almost zero in the bottom waters. Spatial and temporal variations of these variables indicated that the middle zone (0.5 and 1.5 m) was the stratum with a critical role in the functioning of the lake. Indeed, it represented the transition zone between the trophogenic upper water layer (0 to 0.5 m) where photosynthetic activity occurred and the bottom water layer (1.5 to 2.5 m) formed by the microaerophilic to anaerobic tropholitic layer where reducing processes were important.

The rotifer and cladoceran communities were relatively diverse, with 41 species collected during the study year (i.e., 26 species of Brachionidae and Trichocercidae and 15 species of Cladocera), of which 16 are reported for the first time in Cameroon. These organisms filter small organic particles (fresh algal detritus and bacteria) and their presence in the lake could be related to the relatively high algal biomass found there. In terms of relative abundance, rotifers appear to be more important than cladocerans. Among the quantitatively dominant species, the most frequent ones were the Brachionidae (Brachionus angularis angularis, B. falcatus falcatus, B. calyciflorus), the Trichocercidae (Trichocerca elongata elongata and T. bicristata bicristata) and the Cladocera (Ceriodaphnia cornuta and Chydorus eurynotus).

Brachionus angularis was the most abundant and frequent rotifer observed during this study. According to several authors, a high abundance of Brachionus can be considered as a biological indicator of more eutrophic waters. This corroborates the hypereutrophic status assigned to the lake through previous studies dealing with bacterioplankton and phytoplankton in this system. Among Cladocera, Daphnia, an important bacterivore, was absent despite the high density of bacteria (~10⁸ cell mL^-1) in the lake. Previous studies in other Cameroonian aquatic systems reported the scarcity of Daphnia and these earlier authors argued that it was probably replaced by smaller sized species, which in the case of the present study would be Ceriodaphnia cornuta and Moina micrura, the only representatives of Daphniidae encountered. Up to 75% of the number of species identified were periphytic species that generally develop in the littoral zone. The massive occurrence of these species in the plankton sample of the lake was related to the high particle content. Abundance generally peaked in the middle zone of the water column, likely due to the availability of resources and the general potential for such zones to provide a refuge from predation.

However, there did not appear to be any correlation between the rotifer and cladoceran communities and physico-chemical variables when the data set was examined as whole, or grouped by seasons. This suggests that the heterogeneity in species abundances was controlled by other factors, among them sampling points, depths and time of sampling. A multivariate test using canonical correspondence analysis suggested that about 3% and 5% of the variance associated with the species abundance heterogeneity could be explained by sampling point and depth, respectively. Up to 36% of this variance was attributed to the sampling months. We therefore hypothesize that the environmental conditions affecting the development and biological cycles of the studied species are highly dependent on the prevalent seasons found in a tropical zone, i.e. the rainy and the dry seasons.

FR :

Cette note technique présente un système automatique de mesure des pertes de polluants agricoles par ruissellement et drainage souterrain pouvant opérer durant la période estivale et la fonte nivale. Le système de mesure du ruissellement est composé d'une trappe à sédiments, d'un déversoir en V et d'un système automatique de mesure du débit et de contrôle de l'échantillonnage. Le système de mesure du drainage souterrain consiste en un auget à bascule, relié à un compteur d'impulsions, et un partiteur de débit. Ces systèmes ont été utilisés sur neuf parcelles expérimentales pour mesurer les masses de contaminants agricoles perdues par ruissellement et drainage souterrain. La hauteur totale de ruissellement durant les deux années (4,6 mm an^-1) a été faible comparée au drainage (227 mm an^-1). Soixante-seize pourcent du ruissellement annuel s'est produit durant la fonte des neiges. Les principaux avantages de ce système sont de pouvoir mesurer simultanément les volumes et les débits de ruissellement tout en prélevant des échantillons d'eau.

EN :

This technical note presents an automatic measurement system for measuring losses of agricultural pollutants in surface runoff and subsurface drainage outflow. This system can be operated during rainfall events and snowmelt periods. The runoff system was composed of a sediment trap, a V-notch weir and an automatic system for flow measurement and sampling. The subsurface drainage system consisted of a tipping bucket, connected to a pulse counter, and a sampling unit. This system was used on nine experimental plots for the measurement of pollutant losses by surface runoff and subsurface drainage. The total depth of surface runoff measured was low for the two years of measurement (4.6 mm y^-1) compared to subsurface drainage (227 mm y^-1). Seventy six percent of the annual surface runoff occurred during the snowmelt period. The principal advantage of this system is to measure runoff volume and flow while taking water samples.

FR :

Le but de la présente étude était d'étudier l'impact de la variation des niveaux d'eau d'un marais d'eau douce (Baie Saint-François, Québec) sur l'évolution des concentrations et des flux d'hydrogène, monoxyde de carbone, méthane et dioxyde de carbone. Une approche originale impliquant l'association d'un gradient de concentration de ces composés sur un profil vertical de 1,5 m au transfert de flux turbulent micrométéorologique fut utilisée pour la détermination des flux. L'étude démontre qu'une hausse du niveau d'eau d'un bassin versant alimentant une zone humide influence les flux de méthane, de monoxyde de carbone d'hydrogène et de dioxyde de carbone. En conditions submergées, le marais émettait du méthane et du monoxyde de carbone et consommait moins d'hydrogène troposphérique. Ainsi, cette étude démontre que des mesures in situ peuvent servir à inférer des scénarios d'impacts possibles des changements climatiques et des variations des niveaux d'eau sur les émissions des gaz à effets de serre dans l'écosystème du fleuve Saint-Laurent.

EN :

Wetlands are known for their great biodiversity and the important carbon reservoir that they represent. Moreover, in the global warming context, these ecosystems represent net sources or sinks for different greenhouse gases depending of their conditions. For instance, flooded conditions favour methane production whereas they prevent hydrogen and carbon monoxide soil consumption. Baie Saint-François is a freshwater wetland that opens onto Lake Saint-Pierre (St. Lawrence River) where water levels are subject to important fluctuations due to natural processes and human activities (hydroelectricity and navigation). This study was done in order to assess the impact of the Lake Saint-Pierre water level variations on the tropospheric methane, carbon monoxide, hydrogen and carbon dioxide dynamics over the wetland. Knowledge of these dynamics should provide indications about the possible effects of the decreasing or increasing water level associated with the global warming on the production or consumption of these trace gases.

Studies were carried out between June and August 2003 in Baie Saint-François where soil was subjected to successions of flooded and dry conditions. Water and carbon dioxide fluxes were obtained with a Bowen ratio micrometeorological station including a high frequency single infrared gas analyser. Hydrogen, carbon monoxide and methane fluxes were estimated with the modified Bowen method, their vertical concentration gradients (1.5 m) were measured over the plant canopy. The Bowen Ratio station was equipped with different probes to measure parameters such as net radiations, soil heat fluxes and vertical temperature gradients. The turbulent transfer coefficient (k) obtained every 20 min was assumed equal for heat, water vapour and trace gases. Hence, fluxes calculations were done by the multiplication of the turbulent transfer coefficients with the vertical concentration gradients of hydrogen, carbon monoxide and methane.

The instrument used to detect hydrogen, carbon monoxide and methane was a RGA5. This analyser has two detectors: the reductive gas detector (RGD) for hydrogen and carbon monoxide and a flame ionisation detector (FID) for methane. The RGD contains an HgO bed wherein oxygen reacts with reductive gases resulting in Hg° releases detectable by differential UV absorbance. Chemicals were detected continuously in 10 min cycles with an analytical reproducibility of ±0.2, 0.3 and 2% for hydrogen, carbon monoxide and methane. Generally, vertical concentration gradients measured were greater than these limits. A calibration gas containing hydrogen, carbon monoxide and methane at 4940, 1000 and 1000 ppbv respectively in nitrogen was analysed daily to verify calibration. To ensure data integrity, linearity of the instrument was assayed by several dilutions of the standard gas and the integration of the curves gave a correction factor for hydrogen (18%) and carbon monoxide (13%). An intercomparison with NOAA (National Oceanic and Atmospheric Administration) was done to corroborate these correction factors.

Background carbon monoxide, methane and carbon dioxide levels were in agreement with literature values. However, hydrogen was low, as observed by other investigators in summertime, since this season is related to minimal concentrations. Methane followed a diurnal cycle where maximum levels were observed during nighttime. In wet conditions, these nocturnal peaks reached occasionally 4000 ppbv and could be explained by specific production mechanisms and diurnal changes of vertical mixing in the boundary layer. Sensitivity of the processes responsible for methane and carbon monoxide cycling was seen between July 21^st and 26^th where a rain episode (total precipitation of 33.2 mm) increased their background concentrations. It seems that this precipitation was enough to favour methanogenesis and inhibit tropospheric CO and CH₄ consumptions by a reduction of the diffusion of these chemicals into the soil.

Our results demonstrated that four to eleven days following a variation of the Lake Saint-Pierre water level, a change in the tropospheric hydrogen, carbon monoxide and methane concentrations was observed. This lag might be explained by the distance between the lake and the research station (about 1.5 km) and the required time for the adaptation of soil microorganisms to the disruption of their environment. The concentration variations of these chemicals resulted from the inhibition of the processes responsible for their consumption or the activation of the processes accountable for their production.

In June, the wetland was flooded and the CO₂ median flux was -56.5 g m^-2 d^-1. Fluxes increased significantly (Mann-Whitney, α=0.01) in July to 5.30 g m^-2 d^-1, possibly due to dry conditions. Indeed, absence of water favours the activity of soil aerobic microorganisms which might produce more carbon dioxide than the quantity used by plants during photosynthesis.

Methane was produced in June where the median flux was 54 mg m^-2 d^-1. These emissions were caused by the presence of water which maintained anaerobic conditions in the sediments, a suitable environment for methanogenic microorganisms. July was characterised by dry conditions, which generated aerobic environments in soils, an unfavourable microniche for methanogens. Therefore, methane median fluxes decreased significantly (Mann-Whitney, α=0.05) to 0.011 mg m^-2 d^-1 in July. In August, before the end of the investigation period, water levels had increased but methane fluxes were not significantly higher than in July. Moreover, in this period, methane concentrations tended to increase, showing that after an augmentation of the Lake Saint-Pierre water level, Baie Saint-François flooding area could represent a methane source.

During summer 2003, Baie Saint-François acted as a net source of carbon monoxide. In June, the median flux was 21 µg m^-2 d^-1 due to presence of water which inhibited consumption by soil. Emissions were significantly (Mann-Whitney, α=0.05) lower in July (15 µg m^-2 d^-1) due to the absence of water, which represented a suitable environment for microorganisms consuming tropospheric carbon monoxide. In August, the median carbon monoxide flux attained 65 µg m^-2 d^-1 due to an increase of the Lake Saint-Pierre water level. Net carbon monoxide emissions observed in wet and dry conditions might be due to the high organic content in soil and water in addition to the presence of plants since all of these are subjected to photooxidation, generating this pollutant. Therefore, an increase of the Lake Saint-Pierre water level is associated with an augmentation of tropospheric carbon monoxide due to the inhibition of the processes responsible of its consumption.

A decline in the water level might result in the activation of the soil microorganisms (or abiotic hydrogenases) able to consume tropospheric hydrogen. At the beginning of the campaign (June), the median hydrogen flux was weak (-1.37 g m^-2 d^-1) due to the presence of water. However, a net soil consumption was seen in July, where the median hydrogen flux decreased to -125 g m^-2 d^-1. The Lake Saint-Pierre water level increase observed in August was associated with a significant (Mann-Whitney, α=0.05) augmentation of the hydrogen median flux to 299 g m^-2 d^-1. Consequently, a rise in the Lake Saint-Pierre water levels induced an inhibition of the processes responsible of the tropospheric hydrogen consumption.

This study illustrated that the water level fluctuations of the Lake Saint-Pierre have an impact on the H₂, CO, CH₄ and CO₂ dynamics over the surrounding wetlands. When the Lake Saint-Pierre water level decreased, the wetlands acted as a carbon monoxide and carbon dioxide source, but as a consumer of tropospheric hydrogen and a minor source of methane.