Search

Algeria is the largest African and Mediterranean country. It is located in the southern seashores of the Mediterranean Sea. Its climate conditions are ranging from relatively wet to very dry which makes it confronted to high levels of rainfall deficits. The future rainfall evolution may be critical for human activities since increased temperatures may further exacerbate droughts and water shortages. In this study, the regional climate simulations RCA4 are evaluated over historical period 1951–2005 and then used to examine the rainfall and temperature projections over the end of the twenty-first century under two Representative Concentration Pathway (RCP4.5 and RCP8.5) scenarios. The historical simulations are evaluated against observations coming from the recent data sets of Climatic Research Unit (CRU). The trends in precipitation and temperature over historical (1951–2005) and projected future scenarios (2006–2060 and 2045–2100) was depicted by the estimation of the shifts of the three main climate zones existing in Algeria (Koppen-Gieger classification): warm temperate climates (C), steppe climate (BS), and desert climate (BW). Comparative to the mean climate zone surface areas derived from observations (1951–2005), all model simulations predict an expansion of desert climate zone at the expense of the temperate and steppe climate zones. This shift seems to particularly increase by the end of twenty-first century (2045–2100) under RCP8.5 scenario.

The objective of this study is to use two hydrological indices (coefficients of variation and immoderation) to analyze the impacts of dam management methods on seasonal daily flow rate change downstream of three dams: Manouane (diversion-type management method), Ouareau (natural-type management method) and Matawin (inversion-type management method). The results show that this change is far greater downstream of the Matawin dam (characterized by an inversion-type management method) than downstream of the two other dams. Moreover, downstream of the Matawin dam, this daily flow rate change increases significantly over time, while decreasing downstream of the two other dams and in natural rivers. Lastly, this change is better correlated with climate downstream of the Ouareau dam than downstream of the two other dams. It is positively correlated with winter and spring temperatures as well as summer and fall rain. Contrary commonly accepted hypothesis, this study shows that the impacts of dams generally result in an increase of the seasonal flow rate change in Quebec.

The goal of this study was to analyse the impacts of the sequence of 1 El Nino (2009–2010) and 2 La Nina (2007–2008 and 2010–2011) events on the interannual variability of daily streamflow during the growing season (April to September) and grass species abundance downstream from the Taureau reservoir (4,070 km2) on the Matawin River (Quebec, Canada). This reservoir has inverted the natural annual cycle of streamflow: Maximum flows occur in winter and minimum flows in springtime during snowmelt. Comparison of daily flows over the period from 2006 to 2011 using various statistical tests revealed a significant increase in flows released downstream from the reservoir during the 2 La Nina events, with a particularly large increase in the growing season (April to May) during the 1st La Nina event (2007–2008). In contrast, during the El Nino event (2009–2010), streamflow decreased significantly. As far as the abundance of plant species is concerned, the total number of obligate wetland species increased significantly after the 1st La Nina event and then decreased after the El Nino event, along with the total number of terrestrial species. The study shows that relatively intense El Nino Southern Oscillation events can have significant implications for the management of flows released downstream from reservoirs in Quebec and hence affect plant species abundance on islets.

Summary The temporal variability of winter climate in southern Quebec is characterized by increasing temperature and rainfall, but decreasing snowfall. The goal of the study was to analyze the impact of these changes on the magnitude and timing of winter maximum daily flows. Hydrologic series for seventeen tributaries of the St. Lawrence River were analyzed over the period from 1934 to 2010. The most noteworthy result, obtained using the Lombard method, was to highlight a nearly generalized shift in mean values of the series in the early 1970s. This shift reflects a significant increase in the magnitude of winter maximum daily flows after that decade, interpreted to result primarily from snowmelt, which is increasingly frequent at the end of winter. The frequency of the dates of occurrence of winter maximum daily flows increased significantly during the last winter decade (ten days) after the 1970s for most rivers in southern Quebec. The copula method did not, however, reveal any significant change in the dependency between the timing and magnitude of winter maximum daily flows over time. Finally, canonical correlation analysis revealed that the magnitude of winter maximum daily flows is positively correlated with the Pacific Decadal Oscillation (PDO), while their dates of occurrence are correlated with the North Atlantic Oscillation (NAO).

This study presents a comparison of the spatio-temporal variability of characteristics (magnitude, duration and timing) of annual minimum daily extreme flows (AMEF) as a function of land use and the mode of management of dams. Streamflow measured at stations not affected by dams at Joliette, along the L’Assomption River (agricultural watershed, 1340 km2), and at Saint-Michel-des-Saints, on the Matawin River (forested watershed, 1390 km2) on one hand, and downstream from the Rawdon dam (regulated natural-type management mode), on the Ouareau River (1260 km2), which is the main tributary of the L’Assomption River, and from the Matawin dam (inverted-type management mode), on the Matawin River (4070 km2), on the other hand, were compared over the period from 1930 to 2010. As far as the spatial variability of natural rivers is concerned, the magnitude and duration of AMEF are higher in the forested watershed than in the agricultural watershed. In regulated rivers, AMEF magnitude is higher downstream from the dam characterized by a natural-type management mode than downstream from the dam characterized by inversion-type management. However, downstream from the latter, AMEF occur much more frequently and very early in the year. As for temporal variability, the Lombard method did not reveal any influence of land use differences on the stationarity of series of AMEF characteristics. In contrast, differences in dam management mode result in occurrences of AMEF downstream from the inversion-type dam progressively earlier in the year. The duration and timing of AMEF are not correlated with the same climate variables, be it in natural rivers or downstream from dams.

Although numerous studies have looked at the long-term trend of the temporal variability of winter temperature and precipitation in southern Quebec, no study has focused on the shifts in series means and the dependence between these two types of climate variables associated with this long-term trend. To fill these gaps, we used the Lombard method to detect the shifts in mean values and the copula method to detect any change in dependence between extreme (maximum and minimum) temperatures and precipitation (snow and rain) over the periods 1950–2000 (17 stations) and 1950–2010 (7 stations). During these two periods, the shifts in mean values of temperature and precipitation were recorded at less than half of the stations. The only significant change observed at the provincial scale is a decrease in the amount of snowfall, which occurred in many cases during the 1970s. This decrease affected stations on the north shore (continental temperate climate) more strongly than stations on the south shore (maritime temperate climate) of the St Lawrence River. However, this decrease in the amount of snowfall had no impact on the dependence over time between temperature and precipitation as snow.

Although climate models predict that the impacts of climate change on the temporal variability of water levels in the St. Lawrence River will be seasonally-dependent, such a seasonal effect on the current variability of extreme water levels has never been analyzed. To address this, we analyzed the temporal variability of three hydrological variables (monthly daily maximums and minimums, as well as their ratio) of water levels in the St. Lawrence River measured at the Sorel station since 1912, as they relate to climate indices. As for stationarity, the shifts in the mean values of maximum and minimum water levels revealed by the Lombard method took place prior to 1970 for spring water levels, but after that year, for winter water levels. Changes in the winter stationarity are thought to mainly relate to the decreasing snowfall observed in the St. Lawrence River watershed after 1970. In contrast, for spring, these changes are likely primarily related to human activity (digging of the St. Lawrence Seaway and construction of dams). Two shifts in the mean values of fall minimum extreme water levels were highlighted. The first of these shifts, which occurred in the first half of the 1960s decade, can also be linked to human activity (digging of the St. Lawrence Seaway and construction of dams), whereas the second shift, observed after the 1970s for the months of November and December, can be linked to decreasing amounts of snow in winter. AMO (Atlantic Multidecadal Oscillation) is the climate index that is most frequently correlated negatively with the hydrologic variables, mainly in winter and spring.

Underwater light characteristics (quality and quantity) are important drivers of many ecological processes in aquatic systems. However, the mechanisms driving the underwater light climate in large rivers are poorly understood due to the complex interactions among color-producing agents (CPAs), which can vary in space and time. A compelling example of such a complex interaction among CPAs concerns the St. Lawrence River, where chromophoric dissolved organic matter and suspended inorganic particulate matter interact to structure the underwater spectral environment. Because such interactions can be complex, the combined net effect of CPAs on water color is not intuitive and remains to be evaluated. To resolve the spatial and temporal dynamics of the spectral profile in the St. Lawrence River, we analyzed a series of 44 Landsat 5 and 7 images, distributed between 1984 and 2009, which cover the main freshwater section of the river. Rather than following linear trends along the longitudinal axis, the spectral profiles on spectral bands (blue, green and red) presented three distinctive spatial patterns. We argue that matter injected by tributaries and discontinuous zones of the river strongly impacts water color by modulating the balance among CPAs. We also clearly demonstrate a marked trend between 1984 and 2009 toward decreasing reflectance values on bands 1, 2 and 3, presumably in response to changes in land use in the surrounding watershed.

Summary Redundancy analysis (RDA) and Canonical correlation analysis (CCA) were used to examine the complex links between the five characteristics of flow (magnitude, duration, timing, frequency, and variation), defined as part of the “natural flow regime” ecological paradigm, for spring heavy flood, and five climate indices (Atlantic Multi-decadal Oscillation (AMO), Arctic Oscillation (AO), Nord Atlantic Oscillation (NAO), Pacific Decadal Oscillation (PDO) and Southern Oscillation Index (SOI)) in southern Quebec, over the period from 1934 to 2004. The two methods yield similar results. AMO and NAO appear to be the main index correlated with the following four flow characteristics: duration, timing, frequency and coefficient of variation. Thus, positive AMO and NAO anomalies are associated with delayed timing, higher frequency, relatively long duration and relatively weak variability of spring heavy floods in Quebec. This relationship reflects sustained climate conditions characterized by low temperature in winter and early spring in eastern Canada during negative phases of AMO and positives phases NAO. No link was noted between climate indices and the magnitude of spring heavy floods in Quebec.

[1] The relationship between the color (measured with Landsat Thematic Mapper and Enhanced Thematic Mapper sensors in bands 1, 2, and 3) of St. Lawrence River (SLR) water and water levels (in SLR and Lake Ontario) was first analyzed, followed by the analysis of the relationship between water levels and four climate indices for the period from 1984 to 2009. Although there is a statistically significant relationship between these first two variables, this link depends on the strength of two factors: the type of color (band) and inflows from tributaries. Bands 1 (blue) and 2 (green) were negatively correlated with water levels. This correlation, however, is spatially influenced by water inflows from tributaries. Band 3 (red) showed no significant correlation with either water level. Over the same period, water levels were negatively correlated with the Atlantic multidecadal oscillation index. Results suggest that the color of waters in the SLR may be used as an indicator to monitor environmental and climate changes in its watershed.

The goal of this study is to demonstrate the usefulness of the Lombard method for analyzing dam-induced hydrologic impacts. The method was used to accurately detect the effects of the construction of a diversion dam in 2003 on annual and seasonal maximum and minimum daily flows in the Manouane River, Quebec, Canada, measured from 1980 to 2014. The Lombard method yields results that are nearly identical to results obtained using the monitoring (Kruskal-Wallis test) and long-term trend (Mann-Kendall test) methods. The Lombard method revealed a shift in mean values of annual and seasonal minimum daily flows in 2003, the year the dam was built. This shift is sharp for all four seasons. The dam induced a significant decrease in minimum daily flows in all four seasons. As far as maximum daily flows are concerned, unlike the monitoring method, the Lombard method detected a significant decrease only in the mean values of annual and spring maximum daily flows. This decrease occurred two years prior to the construction of the diversion dam. Instead, this decrease is interpreted to be the result of a significant decrease in spring precipitation after 1997. These hydrological changes are different from those induced by other types of dams in Quebec.

We compared the spatiotemporal variability of temperatures and precipitation with that of the magnitude and timing of maximum daily spring flows in the geographically adjacent L’Assomption River (agricultural) and Matawin River (forested) watersheds during the period from 1932 to 2013. With regard to spatial variability, fall, winter, and spring temperatures as well as total precipitation are higher in the agricultural watershed than in the forested one. The magnitude of maximum daily spring flows is also higher in the first watershed as compared with the second, owing to substantial runoff, given that the amount of snow that gives rise to these flows is not significantly different in the two watersheds. These flows occur early in the season in the agricultural watershed because of the relatively high temperatures. With regard to temporal variability, minimum temperatures increased over time in both watersheds. Maximum temperatures in the fall only increased in the agricultural watershed. The amount of spring rain increased over time in both watersheds, whereas total precipitation increased significantly in the agricultural watershed only. However, the amount of snow decreased in the forested watershed. The magnitude of maximum daily spring flows increased over time in the forested watershed.

During the first half of the twentieth century, many hydroelectric facilities were built in the Saint-Maurice River watershed, followed by other human activities in the second half of the century (pleasure boating, boom dismantling, urbanization, etc.). The goal of the study is to constrain the effects of these various types of human activities, particularly those of the many dams in the watershed, on the hydromorphological evolution of the Saint-Maurice River downstream from the La Gabelle (dam) power plant (43,000 km2). Comparison of specific discharge in this river with streamflow measured in a natural river setting reveals a significant decrease in seasonal maximum flows, aside from winter, when daily maximum flows increased significantly. Also, unlike natural rivers, the long-term trend in spring flows is not characterized by a significant change in mean downstream from the La Gabelle plant. These hydrological changes are linked to the inversion-type management mode of the reservoirs built downstream from the plant. As for the morphological evolution, the longitudinal variability of bankfull width downstream from the plant shows two significant shifts in mean: the first, which was quasi-abrupt, took place downstream of the des Forges rapid; and the second, which was gradual, occurred upstream from the confluence of the Saint-Maurice River with the St. Lawrence River, above the point where the Saint-Maurice splits into two branches. Comparison of aerial photographs taken at various times (1948, 1964, 1975, 1996, and 2008) reveals no significant change in the mean of bankfull width over time. However, a significant increase in the surface area of islets located at the confluence was observed, which is caused by sediment accumulation. These sediments were likely derived from local bank erosion resulting from anthropogenic changes.

Comparison of the interannual variability of five characteristics (magnitude, duration, timing, frequency, and variability) of spring heavy floods was carried out for 17 natural rivers in Quebec for the period from 1934 to 2004 to detect any effect of climate warming on these characteristics. This was done using the Lombard method and Copula. Changes in the mean and variance of all characteristics of streamflow were observed, and all these changes are abrupt. Whereas little significant change was observed in the magnitude, duration and variability (CV) series of spring flood flows. A significant change was noted in the frequency (diminution) for five rivers and in the timing of spring floods for six rivers. However, the change in mean timing is the only one that has a hydroclimatic significance in time and space. This change was observed, on the one hand, in the Eastern hydrological region, located on the south shore of the St. Lawrence River, north of 47°N, and characterized by a maritime climate and, on the other hand, in the South-west hydrological region, located on the north shore and characterized by a continental climate. In both cases, the change took place after the second half of the 1970s and is characterized by the early occurrence of spring floods. In addition, in both hydrological regions, the timing of spring floods is correlated with the same hydroclimatic indices, showing a positive correlation with the North Atlantic Oscillation (NAO) and a negative correlation with the Atlantic Multidecadal Oscillation (AMO). Frequency is the only characteristic for which mean and variance changed significantly over time for the same rivers (4), all located north of 48°N, on both shores of the St. Lawrence. For all flow characteristics, the change in variance can predate, postdate or be synchronous with the change in mean. Finally, the dependence between the various characteristics of spring floods as determined using the Kendall tau statistic remained constant in time for most of the rivers.

Built in 1930 on the Matawin River (5770 km²) to supply hydroelectric plants located downstream on the Saint-Maurice River, the Taureau Reservoir caused an inversion of the natural hydrological cycle: maximum flows occur in winter and minimum flows occur in spring during snowmelt. This inversion induces a very rapid rise of the reservoir level (base level for the channel upstream from the reservoir) in April–May during the filling-up process. The goal of the study is to analyze the effect of the rapid rise of the base level on the morphological evolution of the 7-km-long sandy section of the Matawin River located immediately upstream from the reservoir and to compare it with Jiongxin's general model. Comparison of aerial photos taken before (1928) and after (1937, 1950, 1965, 1975, 1983, and 1995) reservoir construction reveals no significant change in the bankfull mean width, sinuosity, and bankfull depth of the studied section. It follows that the morphological evolution of the Matawin River channel upstream from the dam is not consistent with the Jiongxin model. The effect of the inverted hydrological regime in fact produces high channel stability. This stability will probably contribute to a slow but progressive reduction of the channel width and an increase of the sinuosity. This morphological evolution seems typical of the semialluvial streams found upstream from the inverted-type reservoir of the Canadian Shield.

The aim of this study was to test three main hypotheses about the interannual variability of streamflow downstream from dams: (1) an almost similar long-term trend in interannual variability, (2) low variability of flow, and (3) its independence (no link) from climate variability. To test these hypotheses, the interannual variability of winter and spring streamflow downstream from three reservoirs (Gouin, Manouane, and Matawin) which induce an inversion of the natural cycle of streamflow (maximum flows in winter and minimum flows in spring) was compared to the interannual variability of streamflow in natural rivers (measured at the Matawin and Vermillon stations) over the period from 1932 to 2008 in the St-Maurice River watershed. As far as the interannual variability of flow is concerned, its long-term trend is not homogeneous downstream from the three reservoirs in both seasons. However, downstream from two reservoirs, changes in streamflow were observed to be different from those in natural rivers (no significant trend downstream from the Taureau reservoir, on the Matawin River, and significant decrease in spring flow downstream from the Manouane reservoir). Finally, coefficient of variation values for minimum flows are higher downstream from reservoirs than in natural rivers, despite the fact that watershed surface area is larger for regulated rivers than for natural ones. As for the link with climate variability, analysis of the correlation between climate variables (temperature and precipitation) and mean winter and spring daily streamflow reveals that winter streamflow downstream from the Taureau reservoir is not correlated with any climate variable, whereas spring streamflow is positively correlated with rainfall and negatively correlated with maximum temperature. Thus, downstream from reservoirs, the interannual variability of streamflow depends on climate during the spring, but not during winter.

The interactions between precipitation, streamflow and groundwater are very complex. In cold temperate regions characterized by harsh winters, winter streamflow is mainly derived from aquifers that are recharged in the spring, during snowmelt, and in the fall, when evapotranspiration is subdued. Despite this complexity, the modes and trends in the interannual variability of spring (April, May, June and July) streamflow and fall (August, September, October and November) precipitation and streamflow were compared to the modes and trends in the interannual variability of winter (December, January, February and March) streamflow in southern Quebec from 1950 to 2000. Results indicate that the variability modes are identical for all four of these hydro-climatic variables: two modes (south-east and east modes) on the south shore of the St. Lawrence River on either side of the 47°N and a single mode (south-west mode) on the north shore. As for the trend, a significant increase in winter streamflow was observed on the north shore. This increase is comparable to that observed in spring streamflow, which suggests that winter streamflow on the north shore is mainly derived from groundwater recharge during the spring. Moreover, both spring and winter streamflows are positively correlated to the North Atlantic Oscillation climate index. On the south shore, south of the 47°N, a significant decrease was observed in the trend of the interannual variability of winter streamflow, this in spite of a significant increase in fall precipitation and streamflow. An increase in evaporation (decreased infiltration) due to a shift from forest cover to agricultural land cover in this region could account for this. However, fall precipitation and streamflow and winter streamflow are significantly correlated to the Atlantic Multidecadal Oscillation winter index. This correlation is negative with the first two variables but positive with the third. This study suggests that, in southern Quebec, the interannual variability of winter streamflow is mainly affected by spring recharge in non-agricultural catchments (east and south-west modes) and by farming in agricultural catchments (south-east mode). Copyright © 2011 John Wiley & Sons, Ltd.

Winter and summer 7-day low flows of eighteen natural rivers in the St. Lawrence River watershed of Quebec were analyzed over the period from 1934 to 2000. The rivers were first subdivided into three modes using principal component analysis. Two of these modes are located on the south shore of the St. Lawrence River, respectively south (Southeast mode) and north (East mode) of the 47°N, and the third (Southwest mode) is located on the north shore. The Southeast mode shows a significant increase in 7-day low flows in winter and summer due to increasing summer and fall precipitation, whereas the Southwest mode shows a significant decrease in summer 7-day low flows due to an increase in evapotranspiration. No significant change in winter and summer 7-day low flows is observed in the East mode. Results show that the variability of 7-day low flows and their link with climate indices are mainly seasonally dependent. The North Atlantic Oscillation (NAO) is the only index which is correlated to 7-day low flows in all three modes. This correlation is positive for north shore rivers and negative for south shore rivers. Furthermore, only north shore rivers are correlated to NAO in both winter and summer.

Summary We analyzed the temporal variability modes of annual maximum flow of 15 pristine tributaries of the St. Lawrence River in Quebec during the 1934–2000 period in relation to six climatic indices, using principal component analysis, linear regression, the Mann–Kendall test, correlation analysis and canonical correlation analysis. As for temporal variability modes, we identified three variability modes: two on the south shore (on both sides of 47°N) and one on the north shore. A statistically significant temporal trend (positive trend) was observed in the flood time series of the north shore mode only. This increase appears to be due to the continental nature of the climate, which features a strong contrast between the rainy decades (positive anomaly) and the dry ones (negative anomaly). The former (the 1970s and 1980s) followed the latter (the 1940s, 1950s, and 1960s). However, the three modes were characterized by two synchronous anomaly periods on a decadal scale: a negative anomaly during the 1950s and a positive anomaly during the 1970s. That said, these two synchronous periods were characterized by greater amplitude on the north shore than on the south shore. Thus, the dry (wet) periods were dryer (wetter) on the north shore than on the south shore. As for the relationship between the climatic indices and streamflow, Atlantic Multidecadal Oscillation (AMO) was correlated to the north shore mode. ENSO (El Nino/Southern Oscillation Index) was correlated to the south shore mode located north of 47°N) and AO (Arctic Oscillation) was correlated to the south shore mode located south of this parallel. This study shows that floods are not influenced by the same climatic indices on both shores of the St. Lawrence River in Quebec.

We analyzed the temporal variability of annual precipitation measured at five stations in the Macta (14,380 km2) and Tafna (4,949 km2) catchments located in northwestern Algeria, in connection with climate indices over the 1950–2004 period. Analysis of the temporal variability using linear regression and Pettitt test revealed a significant decline in annual rainfall in the mid-1970s in both basins. Furthermore, a decrease of at least 20% of total annual rainfall is also observed at all five stations. As for the relationship between climate indices and variability of annual precipitation, canonical correlation analysis shows that, from 1950 to 2004, precipitation at the five stations are negatively correlated with NINO4 and, to a lesser degree, with NAO.

This paper, based on an analysis of the mean annual flow (MAF) of 16 natural rivers over the period 1934–2000, has the following goals: (1) to determine the different temporal variability modes of the MAF in the St. Lawrence watershed, (2) to describe the temporal variability of the streamflow in each mode and (3) to analyze the influence of the AMO (Atlantic Multidecadal Oscillation) on the interannual and interdecadal variability of mean annual flows in this Quebec watershed. This paper shows that the interannual variability of mean annual flow was not synchronous on both sides of the river. During the period analyzed, MAF variability was characterized by a tendency to decrease on the south shore (right bank) but to increase on the north shore (left bank). A correlation analysis reveals that the influence of the AMO was limited exclusively to the north shore of the river, which is characterized by a continental climate. This correlation is negative. On the south shore, streamflow is correlated negatively to the AO (Arctic Oscillation).

Summary We have compared the monthly flow characteristics between natural rivers (76 stations) and reservoir-regulated rivers (25 stations) based on watershed size, using regression analysis on the monthly flows observed in Quebec. The flow characteristics, defined according to the “natural flows regime” paradigm, are as follows: the periods of occurrence of the monthly maximum and minimum flows and their inter-annual variability, the magnitude and its inter-annual variability, the frequency, and the coefficients of skewness and kurtosis. This study revealed that dams alter all monthly flow characteristics but the extent of these modifications is variable. Two factors influence this extent: the watershed size and the season. The behaviour of large watersheds (>10,000 km 2 ) seems to be different from that of small and medium-sized watersheds. Large watersheds are characterized by a greater inter-annual variability of the timing of maximum monthly discharges but a lower inter-annual variability of the timing of the minimum monthly flows than smaller basins. Concerning the monthly winter flows, large watersheds also exhibit an increase in the magnitude and a smaller variability in between years of discharges. Concerning the season, the hydrological changes were mainly observed in winter and spring. However, in both seasons, the extent of these hydrological changes varies from month to month. These results show that the dam-induced hydrological changes render the method of estimating minimum flow requirements in winter and spring inapplicable to reservoir dams in Quebec to protect Quebec fish habitats effectively. In winter, the method would underestimate the flows released downstream from the dams, while overestimating them in the spring. Moreover, the method cannot be applicable to all watersheds in certain months, because the hydrological changes during those months only affect either small watersheds or large watersheds.

We have analyzed relationships between environmental factors (physiographic, climatic, geological, geographic and land use) and extreme minimum flow characteristics. We have defined 18 hydrological variables on the annual minimum flow series according to the "natural flow regime" paradigm. Application of the principal component analysis (PCA) to these 18 variables allowed us to extract five principal components, each associated with a fundamental characteristic (magnitude-frequency, timing, inter-annual variability of magnitude and timing, shape of distribution curve) of the annual minimum flows. The analysis of indices and values of the loading of the five PC revealed a spatial difference in the annual minimum flow characteristics between stations on the north and south shores of the St. Lawrence River (Quebec, Canada). South shore stations are generally characterized by lower magnitude, later timing, and sharper inter-annual variability of magnitude and timing of the annual minimum flows than those on the north shore. The analysis of correlations (simple correlation and canonical correlation) between the five hydrological PC and the seven environmental PC, also extracted by means of the PCA, revealed that the spatial difference between the annual minimum flow characteristics observed between the north and the south shore stations is mainly due to precipitation (climatic factor) based on a highly complex relationship, which mainly involves the timing of annual minimum flows.

Despite the presence of numerous dams in Qu´ ebec, no study has yet been devoted to their impacts on flood levels. To compensate for this deficiency, we have compared the impacts of dams on the five characteristics (magnitude and its interannual variability, timing and its interannual variability, and asymmetry) of the maximum annual flows between natural rivers and regulated rivers by means of several statistical approaches (analysis of variance, chi-square test, nonparametric tests, etc.). In the course of this study, we analysed 88 stations on pristine rivers and 60 stations on regulated rivers. The latter group was subdivided into three regulated hydrologic regimes, i.e. inversed flow regimes (25 stations), homogenization flow regimes (15 stations) and natural-type flow regimes (20 stations). The following observations emerge from this study. (1) In inversed and homogenization flow regimes, generally associated with reservoirs, all the flow characteristics are modified. These modifications notably entrain a decrease in magnitude, a significant reduction in the frequency of the maximum annual spring flows when the snow is melting and an increase in skewness of the distribution and interannual variability of the magnitude and dates of occurrence of the annual maximum flows. We also observed the disappearance of most flows with a recurrence of over 10 years. All these changes particularly affect watersheds larger than 10 000 km 2 . (2) In natural-type flow regimes, often associated with run-of-river dams, very few changes were observed compared with pristine rivers. These changes primarily affected watersheds smaller than 1000 km 2 . Copyright  2006 John Wiley & Sons, Ltd.

Experiments with marked pebbles were carried out on different sized rivers of the Belgian Ardenne (catchment areas varying from less than 1 km2 to 2700 km2). Specific stream power required to cause bedload movement was evaluated and critical values were obtained. Three types of relationship between critical specific stream power (ω0) and grain size (D) were established. The values for ω0 in the largest river (the Ourthe) were the lowest and were close to the values obtained for mountainous rivers carrying large boulders. In medium sized rivers (catchment area between 40 and 500 km2), the critical unit stream power was higher. It is likely that it is due to the bedform's greater resistance. This resistance would use up some of the energy that can cause movement and transport of bedload. The amount of resistance of the bedform can be expressed as bedform shear stress (τ″), determined by the relationship between grain shear stress (τ′—that determines movement and transport of the bedload) and the total shear stress (τ). This ratio varies between 0.4 and 0.5 in the medium sized rivers, compared to 0.7 in the Ourthe. In headwater streams (less than 20 km2), there is greater loss of energy due to bedform resistance (τ′/τ

The Butgenbach dam on the Warche River was built in 1932 in order to maintain a sufficient supply of water to the Robertville reservoir situated 7 km downstream, for the production of hydroelectricity. During winter months, releases are made almost every day from the Butgenbach dam. From a hydrological point of view, this has resulted in significantly reducing the number of discharges that are higher than bankfull. Despite the reduction in peak discharge, there is a significant increase in the number of efficient discharges (0·6 bankfull). The impacts of these hydrological modifications on the bed morphology and sedimentology below the Butgenbach dam have been studied and the following geomorphological modifications have been identified: a doubling of the width of the channel in 45 years, a reduction in the number of riffles and pools, an increase in the number of gravel bars and islets and an increase in bedrock outcrops in the channel. Moreover, the finest bed particles are mobilized by the almost daily releases, inducing a significant increase in bed-material size sorting. The reduction of sinuosity and the disappearance of bed differentiation and riffle/pool sequences have produced a diminution of bed roughness and an increase of the competence of the river. Thus relatively small floods can remove the armoured layer. Copyright © 2003 John Wiley & Sons, Ltd.

Bed-load transport experiments have been conducted in a steep gravel-bed open ditch. This initially straight ditch has been neglected for many years and looks at present like a second-order natural stream channel. The channel flows through a spruce forest and several log-jams have produced chutes and pools, creating supplementary roughness. The total shear stress has been evaluated using the slope-hydraulic radius product, and the ratio between grain and bed-form shear stresses has been calculated using different methods. The shear stress has also been evaluated from the shear velocities, and this gives a good evaluation of the grain shear stress. Additional experiments have been conducted with marked pebbles to estimate particle mobility and to improve the motion equations. Equations such as θc = a(Di/D50)b)b, defined by Andrews, apply in these cases but the values of a and b are lower than those produced by this author. In a second stage of experimentation, we have destroyed the log jams resulting in a diminution of the roughness and critical shear stress (when the total shear stress is used), an increase of the grain shear stress, and thus greater bed particle mobility for the same discharge. It emerges from these experiments that the log-jams contribute to the reduction of bed-load evacuation and explain the very weak bed-load discharge measured by a bed-load trap (0.3 t · km−2 yr−1).