Navigation – Plan du site

Above- and belowground responses of Populus nigra L. to mechanical stress observed on the Allier River, France

Réponses des parties aérienne et souterraine de Populus nigra L. aux contraintes mécaniques observées sur la rivière Allier, France
Borbála Hortobágyi, Dov Corenblit, ZhuQing Ding, Luc Lambs et Johannes Steiger

Résumés

Des arbres riverains pionniers, tels que le Populus nigra L., établis sur des bancs alluviaux au sein de corridors fluviaux dynamiques, influencent fortement la géomorphologie fluviale en piégeant des sédiments et en construisant des formes fluviales pendant les crues. L'effet ingénieur (changements dans l’état physique de l’habitat par les organismes) de P. nigra sur les bancs alluviaux dépend de sa biomasse et de son degré d’exposition à la contrainte mécanique. P. nigra a une plasticité phénotypique élevée qui lui permet d'adapter ses traits morphologiques et biomécaniques en fonction des conditions hydrogéomorphologiques locales. La compréhension et la quantification de la variation des traits de réponses morphologiques et biomécaniques au sein des populations de P. nigra en fonction de leur degré d’exposition à la contrainte mécanique est fondamentale pour mieux comprendre pourquoi et comment les plantes influent sur la géomorphologie. Dans une étude empirique in situ, nous avons quantifié la relation entre les traits morphologiques et biomécaniques de réponses de P. nigra et son exposition à différents niveaux de contrainte mécanique. En contexte très exposé, en tête de banc, les plantes ont développé des traits de réponse favorisant une plus grande résistance, notamment une tige flexible et basse et un système racinaire plus robuste. En contexte moins exposé, tel le cas de la queue de banc, les plantes ont développé des tiges plus longues et moins flexibles et un système racinaire plus fin. Les plantes établies dans le tronçon aval du chenal de chute alluviale ont développé certains traits communs avec la population de queue de banc et d’autres traits communs avec la population de tête de banc. Les peupliers établis en queue de banc ont favorisé la bioconstruction et donc potentiellement une déconnexion plus rapide des perturbations hydrogéomorphologiques. Nos résultats suggèrent que les processus biogéomorphologiques se produisant à l'échelle fine influent sur les processus à plus large échelle. Ce résultat pointe la nécessité de développer des études biogéomorphologiques hiérarchiques selon des approches ascendante (bottom-up) et descendante (top-down) pour une meilleure compréhension des écosystèmes fluviaux biogéomorphologiques complexes.

Haut de page

Notes de la rédaction

Article soumis le 01 mars 2017, reçu sous sa forme révisée le 14 juin 2017 et définitivement accepté le 01 septembre 2017.

Texte intégral

B. Hortobágyi received funding for her PhD (“Contrat de recherche”) from the French Ministry of National Education, Higher Education and Research (Ministère de l'éducation nationale, de l'enseignement supérieur et de la recherche).

1. Introduction

1Pioneer riparian trees such as Populus nigra L. which establish on gravel bars within dynamic riparian corridors strongly influence fluvial geomorphology by trapping sediments and constructing landforms during floods (Corenblit et al., 2007; Gurnell, 2014; Hortobágyi et al., in press). Ecosystem engineers (Jones et al., 1994) are species which can significantly modify geomorphic processes, landforms and habitats, thus control the availability of resources. Hortobágyi et al. (in revision) showed on a laterally dynamic section of the Allier River, France, that after one to two years of growth, P. nigra saplings significantly influence sediment dynamics on gravel bars because they reached a sufficient biomass. This stage was defined as the “biogeomorphic feedback window” because the riparian plants and geomorphic processes strongly interact in a reciprocal way (Hortobágyi et al., in revision). The authors showed that P. nigra populations which established on alluvial bars under low to intermediate levels of exposure (e.g. bar-tails) to mechanical stress caused by water flow and sediment transport developed a strong aerial biomass and efficiently trapped fine sediments. At the most exposed locations of alluvial bars (i.e. on bar-heads) P. nigra populations were able to resist mechanical stress but they did not develop an important aerial biomass, and thus they did not trap significant quantities of fine sediment. These results suggested that the morphological and biomechanical “traits” and related “attributes” of young P. nigra populations which colonize alluvial bars, change according to variations in exposure to mechanical stress (i.e. shear stress imposed by water flow; sediment transport) and that these changes result in varying aptitudes of the plants to trap fine sediments during annual floods. Plant traits are morphological, biomechanical, physiological and phenological characteristics of plants which can be measured at the individual level; the value or the modality taken by a trait is called an “attribute” (Violle et al., 2007). When the attribute of a trait varies in response to changes in environmental conditions (e.g. water flow, sediment erosion, transportation, deposition), the trait is called a “response trait”. Traits which affect environmental conditions (e.g. flow characteristics, geomorphic landforms), community or ecosystem properties, are considered as “effect traits” (Violle et al., 2007; Corenblit et al., 2015).

2P. nigra has a strong phenotypic plasticity, i.e. individuals can adapt their morphological and biomechanical traits according to the local hydrogeomorphic conditions (Karrenberg et al., 2003; Chamaillard, 2011; Corenblit et al., 2014). Such plasticity improves P. nigra individual’s probability to establish under harsh environmental conditions (e.g. exposed to flood disturbances) and eventually to reach their biogeomorphic feedback window (Eichel et al., 2015; Hortobágyi et al., in revision). During establishment (i.e. first three years following germination), the sapling’s growth pattern is highly controlled by local hydrogeomorphic conditions, such as sediment texture, topographic level, hydrological regime and exposure to mechanical stress that act as a strong ecological filter (Cooper et al., 1999). Different authors have suggested that in order to increase their resistance to mechanical stress, saplings develop specific morphological and biomechanical attributes (Karrenberg et al., 2002; Bornette et al., 2008; Puijalon et al., 2011). Variation in trait attributes is supposed to be dependent upon the level of mechanical stress with a trade-off related to the function of anchorage and resource acquisition (Karrenberg et al., 2003; Read and Stokes, 2006; Pasquale et al., 2013).

3The quantification of morphological and biomechanical response trait attributes variation in P. nigra populations according to their location on alluvial bars is essential to understand why these ligneous riparian plants are capable to impact fluvial morphodynamics and fluvial landforms. To our knowledge, very little is known about the in situ aboveground and belowground morphological and biomechanical response of saplings in their early stage of development on alluvial bars with the exception of very recent studies (Kui and Stella, 2016). We hypothesise that (i) when highly exposed to shear stress, saplings predominantly develop functional morphological and biomechanical traits increasing their resistance to uprooting, such as a reduced size and a strong flexibility which limits drag force, i.e. “avoidance traits”, and also other traits such as a strong root system that increases anchorage, i.e. “tolerance traits”; (ii) conversely, under less exposed situations, their morphology will predominantly be the expression of the function of resource acquisition, i.e. water and nutrient uptake, with varying root/shoots ratios but a weaker structural root system (the roots with a smaller diameter) and potentially a taller above-ground size under good growth conditions.

4In this empirical in situ study, we will focus on the early stage of the widely (on the European continent) distributed P. nigra which is also an abundant species on the alluvial bars of the laterally dynamic Allier River, France. The main objective was to explore if contrasted morphological and biomechanical responses exists in-between the P. nigra populations in relation with exposure to mechanical stress. Three contrasting locations on alluvial bars were distinguished: (i) on the most exposed upstream location of the alluvial bar, hereafter called bar-head; (ii) on the less exposed downstream location of the alluvial bar, hereafter called bar-tail; and (iii) within the lower reach of a chute channel which is sheltered during low annual floods but more exposed to concentrated water flow during more important floods.

2. Methods

2.1. Study site and location of P. nigra populations

5Populus nigra L. individuals were sampled in spring 2014 within a reach of the lower gravel bed Allier River, France, near Châtel-de-Neuvre (fig. 1) which is evolving from a transitional wandering style to a meandering style. This river reach within the “Réserve Naturelle Nationale du Val d’Allier” with a certain protection status and which experienced moderate anthropogenic impacts, is characterized by an active lateral erosion in the outer bends of meanders and point bar formation and migration in the inner bends (Petit, 2006; Dejaifve and Esquirol, 2011). The Allier River has a pluvial hydrological regime with strong seasonal and interannual variability and a mean annual discharge of 117 m³.s-1 (1986-2017; data: Banque Hydro http://www.hydro.eaufrance.fr). P. nigra individuals were sampled within three populations located on two alluvial bars juxtaposed in the downstream direction and within one chute channel. The three populations were mainly composed of P. nigra individuals chosen from the same local populations, however some Salix purpurea L. and Salix alba L. individuals were also present within the sampled vegetation patches.

6These three populations dating back to 2011 and 2012 were chosen in order to (i) obtain three different levels of exposure to mechanical stress with homogeneous sediment texture at the patch scale, (ii) and to maximise the probability that the populations originated from the same local population, and thus dispose of a close genotype. The first population (fig. 1C) is located on the highly exposed bar-head of the downstream alluvial bar; the second population on a less exposed bar-tail location of the upstream alluvial bar (fig. 1D) and the third population within the lower (downstream) reach of the chute channel of the upstream alluvial bar (fig. 1E).

Fig. 1 – Localisation map.
Fig. 1 Carte de localisation.

Fig. 1 – Localisation map.  Fig. 1 – Carte de localisation.

A. Study site; B. Sampling sites on the Allier River. C. Bar-head; D. Bar-tail; E. Lower reach of chute channel
A. Zone d’étude ; B. Sites d’échantillonnage sur la rivière Allier. C. Tête de banc ; D. Queue de banc ; E. Tronçon aval de la chute alluviale.

7The flooding threshold of all three populations corresponds to a discharge of about 130 m3.s-1 (fig. 2). Total submersion time, determined from flow hydrographs, of the two year old P. nigra saplings was about 342 days. During the first year the saplings were exposed to two quasi-annual floods of 357 and 408 m3.s-1 (2‑years return period: 580 m3.s-1) and to one flood of 729 m3.s-1 (5‑years: 810 m3.s-1). During the second year after recruitment, three more floods of 610, 364 and 375 m3.s-1 occurred. The first two floods occurred during autumn and winter when saplings were leafless and the third flood occurred during spring time when leaves had sprouted at the end of the dormancy. The last three floods occurred during spring time, full summer (maximum biomass) and winter (leafless period). Submersion time of one year old P. nigra saplings was around 69 days. These saplings were only exposed to the last three floods.

Fig. 2 – Daily mean discharges (m3.s-1) at the hydrological station of Châtel-de-Neuvre between 2011 and 2014 and possible recruitment periods of P. nigra populations (data. Banque Hydro http://www.hydro.eaufrance.fr; station code: K3400810).
Fig. 2 – Débits journaliers (m3/s) à la station hydrologique de Châtel-de-Neuvre entre 2011 et 2014 et les périodes de recrutement possible des populations de P. nigra (données. Banque Hydro http://www.hydro.eaufrance.fr ; code de station : K3400810).

Fig. 2 – Daily mean discharges (m3.s-1) at the hydrological station of Châtel-de-Neuvre between 2011 and 2014 and possible recruitment periods of P. nigra populations (data. Banque Hydro http://www.hydro.eaufrance.fr; station code: K3400810).Fig. 2 – Débits journaliers (m3/s) à la station hydrologique de Châtel-de-Neuvre entre 2011 et 2014 et les périodes de recrutement possible des populations de P. nigra (données. Banque Hydro http://www.hydro.eaufrance.fr ; code de station : K3400810).

2.2. Field sampling

8First, all selected poplar individuals were localised using a DGPS (Magellan PM500). One surface and one sub-surface sediment sample was taken in order characterize overall sediment texture for each of the three populations. The subsurface sample was taken at the level where a clear transition in sediment texture was detected. Sediment texture was determined at the laboratory using standard sieving procedures (Rivière, 1977).

9Vegetation parameters were measured in situ: the above ground height (cm), diameter (cm) at ground level and at 20 cm, inclination of the stem (degree) and flexibility (Newton). Initial stem inclination was measured along the channel upstream-downstream axis. Higher values than 90o indicate that the stem is bent to the downstream direction. In order to quantify biomechanical properties of the individuals, the stem flexibility was measured with an electronic gauging force (Sauter FH 50; Newton range of 0.01-50): from its initial position, the stem was bent by an additional 10° along the upstream-downstream axis of the flow channel. High values indicated the necessity of applying a strong force to bend the stem, and therefore a low flexibility of the individual. Finally, plants were excavated using conventional shovels in order to measure parameters related to the belowground part of the plants: the length of buried stem (cm), the length (cm) and diameter (cm) of the taproot, diameter of the collar (cm), number of lateral and adventitious roots and the number of roots with a diameter of > 0.3 cm which was chosen because it is suggested that it corresponds to the structural roots (Ding, 2014). Structural roots are more specifically related to the function of anchorage and fine roots to the function of nutrient uptake (Stokes et al., 1995; Read and Stokes, 2006). Between seedling recruitment and field sampling, the river deposited additional sandy sediment on the study sites during high water stages (> 130 m3.s-1). The length of the buried stem was measured between the root collar and the bar or chute channel surface. Roots growing on this part of the stem are adventitious roots, while lateral roots are growing from the original taproot. Additionally, we calculated the ratio between the aerial and the buried parts of the stem and also the total length of the plant (i.e. the above- plus the belowground part on the plant). In spring 2014, we excavated a total of 96 P. nigra individuals within the three populations, (bar-head [pop1]: 46 individuals; bar-tail [pop2]: 40; chute channel [pop3]: 10). From the 96 individuals, 62 individuals (respectively 23, 32 and 7) were excavated entirely without breakage of the tap root. Stems were cut and sanded at the collar, then they were analysed with a binocular microscope and scanned at a high resolution. Annual growth rings were counted with a high precision. This dendrochronological analysis of each individual confirmed that the three populations had the same age, and that a statistical comparison of their morphological and biomechanical responses to mechanical stress could be carried out.

2.3. Data analysis

10We used two datasets for the statistical analyses. The first dataset included all (in total 96), i.e. intact and broken (at tap root level) individuals. This dataset was used to study the aboveground response of P. nigra (pop1: 46; pop2: 40; pop3: 10 individuals). The second dataset included only the 62 intact individuals (pop1: 23; pop2: 32; pop3: 7 individuals). This dataset was used to study the belowground and the complex (above- and belowground) response of P. nigra.

2.3.1. Descriptive statistics

11First, we used descriptive statistics and statistical tests in order to determine if plant response trait attributes varied in relation with the different levels of exposure to mechanical stress (bar-head, bar-tail, chute channel). We used Kruskal-Wallis non-parametric test with Steel-Dwass-Critchlow-Fligner multiple pairwise comparison procedure to identify which sites were responsible for rejecting H0. Letters are assigned to groups (i.e. A, B) indicating that the groups are significantly different. The analysis was undertaken with XLSTAT software.

2.3.2. Discriminant analysis

12P. nigra individuals had known group membership (three distinct locations). After data standardisation, we applied discriminant analysis to explain and predict the membership to several groups. This method allows (i) to check on a two-dimensional chart if the groups are well discriminated; (ii) to describe group properties using explanatory variables; and (iii) to predict to which group an observation belongs. Statistical tests were calculated, such as the Box test (Fisher's F asymptotic approximation) and the Wilks' Lambda test (Rao's approximation). The Box test is used to verify the assumption of equality for intra-class covariance matrices. The Wilks' Lambda test allows to test if the vector of the means for the various groups are equal or not. In addition, the confusion matrix was calculated showing the correctly and incorrectly assigned observations to groups based on the discriminant analysis. We have also undertaken a cross-validation to compensate for an optimistic apparent error rate. In cross-validation each observation one at a time is removed and the classification function is recalculated using the remaining data, and then the forecast is calculated for the omitted observation. The analysis was undertaken for the complete dataset (aboveground and belowground traits) within the XLSTAT software to better understand the response of the entire plant with a possibly correlated response between the aboveground and belowground traits to different levels of mechanical stress.

13In the second part of the analysis, we have undertaken a cartographic analysis to study plant responses within each of the three locations (within-patch analysis). For each observation, its membership class was assigned by the discriminant analysis model, which is deduced by the membership probabilities. We mapped the probability values and the assigned group membership for each observation to determine where incorrectly assigned observations were localised. The cartographic work was undertaken in ArcMap™ using the probabilities and the membership classes of the cross-validation model, which is a more realistic model.

3. Results

3.1. Sediment texture and topography

14The three locations differed according to sediment texture. The coarsest sediment texture was found on the bar-head and the finest on the bar-tail (fig. 3). There was a significant difference between the altitudes of the three populations. The population in the secondary channel was located at the highest mean elevation (219.14 ± 0.04 m), then the downstream population (218.63 ± 0.11 m) and finally the upstream one (218.22 ± 0.21 m). The relative altitudes of the three locations are respectively 1.65 m, 1.13 m and 0.81 m above the water level (frequency of 0.45).

Fig. 3 – Surface and subsurface cumulated particle size distributions of the three sample sites.
Fig. 3 – Courbes granulométriques cumulées des trois sites à la surface et à la subsurface.

Fig. 3 – Surface and subsurface cumulated particle size distributions of the three sample sites.Fig. 3 – Courbes granulométriques cumulées des trois sites à la surface et à la subsurface.

3.2. Age of patches

15The dendrochronological analysis showed that the age of Populus nigra L. individuals within the three patches was comprised between one and three years. The mean age of the three populations including intact and broken individuals was 1.9 at bar-head, 1.6 at bar-tail and 2.3 years in the chute channel. For the dataset including only intact individuals the mean age was respectively 2, 1.6 and 2.2 years. There was no significant difference (tested with Mann-Whitney) concerning the height, the total length and the root collar diameter within groups between individuals of different age. Thus, age differences within populations did not induce a statistical bias for the in-between population analysis of differences related to response traits. A sign of mechanical impact was recorded between the first and the second ring of seven individuals within the upstream part of the bar-head population.

3.3. Populations’ response traits: descriptive statistics

3.3.1. Aboveground response traits

16For all aboveground response traits (height, flexibility, diameter at ground level and at 20 cm and inclination of the stem) the Kruskal-Wallis test rejected the H0. Thus, P. nigra trait attributes significantly varied between the three populations (tab. 1). The height of the plants was significantly different between all three locations (Steel-Dwass-Critchlow-Fligner multiple pairwise comparison) with the tallest individuals located in the chute channel and the smallest within the bar-head population. For all other aboveground traits a significant difference could be detected between the bar-head population and the two others. At the bar-head location, plants had a more flexible stem with smaller diameters and a higher inclination than plants located at the bar-tail location and in the chute channel. However, the sampling size was reduced within the chute channel population and results must therefore be interpreted carefully.

Tab. 1 – Summary statistics and results of Kruskal-Wallis test (α = 0.05) of aboveground and belowground P. nigra traits. Letters assigned to groups (i.e. A, B) are indicating that the groups are significantly different.
Tab. 1 Statistiques sommaires et résultats du test de Kruskal-Wallis (α = 0,05) des traits aériens et souterrains du P. nigra. Les lettres associées aux groupes (i.e. A, B) indiquent que les groupes sont significativement différents.

Tab. 1 – Summary statistics and results of Kruskal-Wallis test (α = 0.05) of aboveground and belowground P. nigra traits. Letters assigned to groups (i.e. A, B) are indicating that the groups are significantly different.  Tab. 1 – Statistiques sommaires et résultats du test de Kruskal-Wallis (α = 0,05) des traits aériens et souterrains du P. nigra. Les lettres associées aux groupes (i.e. A, B) indiquent que les groupes sont significativement différents.

3.3.2. Belowground response

17Concerning the belowground response traits, the Kruskal-Wallis test showed significant differences related to the length of the buried stem, the taproot and root collar diameters and the number of structural roots with a diameter of > 0.3 cm. No significant differences were observed regarding the length of taproot and the number of lateral and adventitious root (tab. 1). The length of the buried stem was more important at the bar-head and bar-tail locations and was significantly shorter in the chute channel. The ratio of aerial and buried stem was close to the one within the bar-tail population demonstrating that the aerial part of the stem is equal to the buried one. Within the bar-head population the buried stem was longer, while in the chute channel population it was shorter than the aerial part. The total length (sum of the aerial, the buried stem and the taproot) of the plant was the longest on the bar-tail and the shortest at the bar-head population. The taproot and root collar diameters were the smallest at the bar-tail location and the largest on the bar-head and in the chute channel. The number of structural roots with a diameter of > 0.3 cm showed the same pattern as the aboveground traits with a high similarity between the bar-tail and the chute channel populations. At these two locations P. nigra individuals exhibited a weaker proportion of structural roots than within the bar-head population.

3.4. Discriminant analysis: complex (above- and belowground) response of P. nigra

18The Box test confirmed that we need to reject the hypothesis that the covariance matrices are equal between the groups and the Wilks' Lambda test confirmed that the difference between the mean vectors of the groups are statistically significant. The two-dimensional chart representing the observations on the factor axes confirmed that the tree populations are well discriminated (fig. 4A). Bar-head and bar-tail populations were discriminated along the first axis, while bar-tail and chute channel populations along the second axis. 100% of the variance is represented by the two factors. The correlation between the initial variables and the two factors are represented on the Figure 4B. The factor F1 was the most correlated with the aboveground height, the inclination and the number of structural roots with a diameter of > 0.3 cm; the factor F2 with the diameter of root collar and taproot and the length of the buried stem.

Fig. 4 – Discriminant analysis based on above- and belowground response traits.
Fig. 4 Analyse factorielle discriminante basée sur des traits aériens et souterrains.

Fig. 4 – Discriminant analysis based on above- and belowground response traits.Fig. 4 – Analyse factorielle discriminante basée sur des traits aériens et souterrains.

A. Two-dimensional chart representing the observations on the factor axes; B. Correlation between the initial variables and the two factors. 1. Bar-head; 2. Bar‑tail; 3. Lower reach of chute channel.
A. Diagramme bi-dimensionnel représentant les observations sur les axes factoriels ; B. Corrélation entre les variables initiales et les deux facteurs. 1. Tête de banc ; 2. Queue de banc ; 3. Tronçon aval de la chute alluviale.

19The upstream population (bar-head) was characterised by a small height, high inclination, a strong abundance of structural roots with a diameter of > 0.3 cm, large root collar and taproot diameter and a long buried stem. The downstream population (bar-tail) was characterised by an important stem height, low inclination, low abundance of structural roots, small root collar and taproot diameter and a long buried stem. The population in the chute channel was characterised by an important stem height, low inclination, strong number of structural roots, large root collar and taproot diameter and a small buried stem. 100% of the observations were well classified and 80.65% of the observations were well predicted by the cross-validation (tab. 2). Only six individuals were classified by the cross-validation as individuals belonging to the downstream population instead of to the bar-head, four as belonging to the bar-head instead of belonging to the downstream population and two as belonging to the downstream population instead of to the chute channel.

Tab. 2 – Confusion matrix for the estimation sample and for the cross-validation.
Tab. 2 – Matrices de confusion pour l’échantillon d’estimation et pour les résultats de la validation croisée.

Tab. 2 – Confusion matrix for the estimation sample and for the cross-validation.Tab. 2 – Matrices de confusion pour l’échantillon d’estimation et pour les résultats de la validation croisée.

20The cartographic results suggest that between-patch, as well as within-patch gradients for plant responses were present (fig. 5). Some individuals of the most exposed bar-head location were assigned to lower exposed bar-tail location and conversely. The misclassified individuals within the less exposed chute channel were assigned to the bar-tail group but not to the bar-head one. The misclassified individuals within the bar-head patch are rather located at the downstream part of this patch and the misclassified individuals within the bar-tail patch are rather located at upstream part of this patch.

Fig. 5 – Membership probabilities and membership classes of observations deduced by the probabilities in cross-validation mode.
Fig. 5 Probabilité d’appartenance à un groupe et classes d’appartenance des observations déduites par les probabilités en mode de validation croisée.

Fig. 5 – Membership probabilities and membership classes of observations deduced by the probabilities in cross-validation mode.  Fig. 5 – Probabilité d’appartenance à un groupe et classes d’appartenance des observations déduites par les probabilités en mode de validation croisée.

A. Membership probability to be assigned to the bar-head population; B. Membership probability to be assigned to the bar-tail population; C. Membership probability to be assigned to the chute channel population. Note that thresholds between classes are different for A, B and C (see legend).
A. Probabilité d’appartenance attribuée à la tête de banc ; B. Probabilité d’appartenance attribué à la queue de banc ; C. Probabilité d’appartenance à être attribuée à la chute alluviale. Notez que les seuils de classes sont différents pour chaque site (voir légende).

4. Discussion

21In accordance with our first hypothesis that within exposed locations young poplar saplings increase their resistance to uprooting, the Populus nigra L. population located on the highly exposed bar-head responded to mechanical stress with a reduced size and a strong flexibility of stems (sections 3.3.1 and 3.4). The stems also showed an inclination in the downstream direction parallel to flow. The development of these two trait attributes related to stem size and flexibility was observed in experimental in situ (Puijalon and Bornette, 2004) and ex situ (Puijalon et al., 2008) studies, for example on Mentha aquatica L. plants in response to mechanical stress related to water flow and for Ulmus americana L. plants subjected to flexural treatment in an ex situ greenhouse experiment (Telewski and Pruyn, 1998). These observed response trait attributes most likely correspond to avoidance trait attributes as defined by Puijalon et al. (2011): the plants reduce their above-ground size to limit mechanical damage and to prevent uprooting. As corollary of the reduced plant size, the limitation of the drag decreases the potential of the plants to trap large amounts of fine sediment. Hortobágyi et al. (in revision) demonstrated on several alluvial bars of the Allier River within the same study reach (including the present bars) that such a limitation of engineer effects caused by mechanical stress in the most exposed locations on bars seems to reflect an inhibiting effect of niche construction by riparian plants. Sediment texture of the bar-head was constituted of coarse sediments, coarser than the chute channel and the bar-tail (fig. 6), suggesting a limitation of surface plant engineer effects and the occurrence of a dominantly physically-driven sedimentation process. Our results suggest that P. nigra plants which establish within exposed bar-head locations are able to resist strong mechanical stress (shear stress and coarse sediment transport and burial) by developing aerial avoidance traits, but they do not significantly contribute to sediment trapping because of their small size and low structural density. A trade-off based on the coast-benefits balance between the function of resisting mechanical constraints and improving resource storage must be found by the plants. Therefore, we argue that in the highly exposed contexts, short term sapling survival is the priority of the individual plant and that engineering effects remain non-significant.

22Within the bar-head, we also observed that the buried part of the stem of all individuals sampled was proportionally greater than the part of the aerial stems (sections 3.3.2 and 3.4). Furthermore, the buried stems lay close to a horizontal position under a coarse sediment layer (fig. 6). These observations suggest a strong biomechanical impact of floods and especially bedload transport on the exposed saplings resulting in a high stem inclination and coarse sediment burial that could lead to the improvement of anchorage of the plants through specific adaptations. Stokes et al. (1995) showed that for trees exposed to wind stress, windward roots (i.e. roots growing towards the mechanical force, here the wind) are playing an important role in stability. We suggest that the buried stems of the observed P. nigra plants and their associated abundant and strong adventitious roots provide an increased anchorage capacity. The high share of buried stems leads to a smaller part of the plant being directly exposed to flow and shear stress and contributes to the further development of adventitious roots and thus to an additional increase of anchorage.

Fig. 6 – Schematic representation of Populus nigra L. exposed to different levels of mechanical stress. The length of adventitious and lateral roots remains unknown since they could not be excavated entirely. Question marks indicate that root lengths are unknown.
Fig. 6 Représentation schématique des Populus nigra L. exposés au stress mécanique de différents niveaux. La longueur des racines adventives et latérales est inconnue, étant donné qu’elles n’ont pas été excavées dans leur intégralité. Les points d’interrogation indiquent le manque d’information sur la longueur des racines.

Fig. 6 – Schematic representation of Populus nigra L. exposed to different levels of mechanical stress. The length of adventitious and lateral roots remains unknown since they could not be excavated entirely. Question marks indicate that root lengths are unknown.Fig. 6 – Représentation schématique des Populus nigra L. exposés au stress mécanique de différents niveaux. La longueur des racines adventives et latérales est inconnue, étant donné qu’elles n’ont pas été excavées dans leur intégralité. Les points d’interrogation indiquent le manque d’information sur la longueur des racines.

23However, within the population of the bar-head, we did not observe any increase in tap root length compared to the two other populations (sections 3.3.2 and 3.4). This is in line with the results of Tamasi et al. (2005), who studied root response of Quercus robur L. plants to wind loading and did not observe significant difference in tap root length between wind loaded plants and the control group. It was suggested that plants are rather investing energy in lateral root growth because they provide the major component of anchorage strength in trees (Stokes et al., 1995). As expected, within the bar-head population, structural roots with a diameter of > 0.3 cm were significantly more abundant, and collar and taproot diameter were more important, than within the population of the bar-tail. This tolerance strategy (sensu Puijalon et al., 2011), which enables plants to reach a higher resistance to uprooting, was also observed in an ex situ glasshouse experiment comparing sunflower and maize seedling responses to mechanical stimulation in the form of stem flexing (Goodman and Ennos, 1996). In addition, the development of such tolerance trait attributes (strong root system) certainly impacts geomorphology through the persistence of P. nigra individuals under highly exposed locations by increasing sediment cohesion, which is a widely recognized role of riparian vegetation (Abernethy and Rutherfurd, 1998; Polvi et al., 2014).

24Adventitious roots that rapidly develop in freshly deposited sediments also improve the capacity of the plants to explore the substrate for water and nutrient uptake. This is especially important on the bar-head within a coarse and rather nutrient poor sediment environment. Therefore, and despite a small aboveground plant size, the adventitious roots contribute through an increase in anchorage and water and nutrient uptake to an improved capacity of the plants to survive under highly disturbed and stressful conditions and to resist mechanical destruction. A further advantage of the development of the adventitious root systems lies in its potential response to changes in geomorphology, e.g. in relation to lateral channel migration or avulsion. Within a less disturbed and stressful environment the earlier developed belowground biomass (long buried stem with adventitious roots) can facilitate rapid aboveground biomass production.

25In accordance with our second hypothesis that under less exposed situations plant morphology predominantly expresses the function of resource acquisition, saplings on the bar-tail and within the lower chute channel showed longer, larger, less flexible and less inclined stems (sections 3.3.1 and 3.4). The total plant length was maximal within the bar-tail population (tab. 1), suggesting that fine sediment burial under less exposed situations stimulates aboveground biomass production. Some woody plant species of coastal dunes (i.e. Populus balsamifera L. and Salix cordata Michx.) also respond with increased vertical shoot growth to burial stress (Dech and Maun, 2006). Thus, the positive growth response we observed in the bar-tail population seems clearly to be linked to the combination of the decrease in exposure to mechanical stress and the occurrence of fine sediment deposition, which provides nutrients and an improved moisture retention capacity during summer (Steiger and Gurnell, 2003; Francis et al., 2009). Furthermore, the correlation between fine sediment trapping and plant morphology attributes was studied in various contexts, such as in a flume experiment on marsh species (Bouma et al., 2013), in a gully erosion project (Erktan and Rey, 2013) and in riparian environment (Euler et al., 2014; Corenblit et al., 2015; Manners et al., 2015). Within the fluvial environment, higher plants that exhibit a larger biomass are expected to induce greater positive topographic changes (Kui et al., 2014; Diehl et al., in press). The buried stems covered by fine sediment observed within the bar-tail population suggest such an effect of the P. nigra individuals on fine sediment trapping. Hortobágyi et al. (in revision) highlighted on the same alluvial bars of the Allier River, that P. nigra engineer effects on sediment trapping are indeed the most efficient when plants are exposed to intermediate and low mechanical stress, and, concurrently, develop a larger aerial biomass. Furthermore, a very high plant density was observed within the bar-tail population leading to a high roughness and thus a very efficient sediment trapping capacity.

26Our results combined with the ones obtained by Hortobágyi et al. (in revision) suggest that within the bar-tail population, changes in P. nigra morphological and biomechanical trait attributes on alluvial bars are not only a passive response to mechanical stress but that they result from induced fine sediment trapping and positive feedbacks between P. nigra plant growth and fluvial landform construction (Corenblit et al., 2007, 2015; Gurnell, 2014). Landform construction, as observed during P. nigra establishment on the Allier River, might be considered as a niche construction strategy leading to a progressive decrease of mechanical stress and modifications of habitat conditions that positively impact P. nigra survival and growth (Corenblit et al., 2014). Furthermore, the niche construction process, as observed during the present study, might be reinforced through the sheltering effect of plants growing within the most upstream and exposed areas of the studied vegetation patch. Corenblit et al. (2016) showed that on the alluvial bars of the channelized Garonne River, France, highly exposed P. nigra cohorts offer bioprotection for younger cohorts which establish just downstream. The sheltering effect of upstream established plants was also demonstrated in relation to the survival of seedlings and cuttings of P. nigra, S. alba and S. elaeagnos (Moggridge and Gurnell, 2009). Our within-site results (fig. 5) suggest that a plant sheltering effect exists from the beginning of establishment, i.e. during the first two years following recruitment, within the same patches composed of individuals of the same age or very close age forming dense stands. In such patches, individuals located downstream benefit from the protective effects of the upstream individuals (i.e. intra-specific facilitation). However, these exploratory results have to be confirmed by repeated sampling within a shorter time period to exclude any differences linked to the development of saplings.

27The population located in the chute channel differed from the two others by its aboveground stem which was longest in the chute channel, and its buried stem which was shortest in the chute channel. The total plant length was shorter compared to the bar-tail population. This result emphasizes aerial biomass growth stimulation provided by fine sediment burial that increases the local potential of water and nutrient acquisition. Individual plants that establish in the lower reach of the chute channel developed some common trait attributes in comparison to the bar-tail population and some other trait attributes which were common to the bar-head population. The long, large, less flexible and less inclined stems and the low abundance of structural roots with a diameter > 0.3 cm were similar with the bar-tail population which was less exposed to mechanical stress. However, the larger collar and taproot diameter are common trait attributes in comparison with the bar-head population. This contrast within the same population may be related to the fact that (i) P. nigra plants growing within lower chute channels benefit from favourable habitat conditions (i.e. water and nutrients pools combined with a decreased exposure to mechanical stress during annual floods) thus favouring biomass production; (ii) they endure more mechanical stress during higher flows and floods because chute channels function as flood channels where water flow converges. The close resemblance concerning the collar and taproot diameter with the bar-head population might also be explained by the similarity in sediment texture (coarse sediment). An additional explanation which could not be further explored in the present study was given by Ennos (1993) who observed that plants compensate their aboveground growth to develop an optimal anchorage system to its aboveground biomass.

28Overall, our observations indicate that P. nigra populations can develop a set of variable morphological and biomechanical responses to contrasted hydrogeomorphic constraints, leading to functional diversity and an increased resistance of the plants at the scale of an alluvial bar. At the highly exposed bar-head location, plants rather developed avoidance and tolerance response traits without significantly affecting sediment trapping while at less exposed bar-tail locations effect traits seem to be operant resulting in a positive feedback between plant growth and fluvial landform construction.

5. Conclusion

29In this biogeomorphic study we provided a quantitative understanding of the relation between response trait attributes of Populus nigra L. plants and their exposure to different levels of mechanical stress caused by flow and sediment transport. At the highly exposed location, plants developed both avoidance and tolerance response traits allowing a higher resistance to mechanical stress, while at the less exposed location plants developed response trait attributes allowing a better resource acquisition and higher growth rate thus favouring bioconstruction. These results support the hypothesis of a positive niche construction by riparian poplar plants within fluvial corridors (Corenblit et al., 2014). Our study, in addition to the one undertaken by Hortobágyi et al. (in revision), suggests that differences in plant response trait attributes and their related effects on fluvial landform construction may lead to different biogeomorphic evolutionary trajectories on alluvial bars from a functional trait perspective. These findings highlight the importance of considering in more detail the role of fine scale biogeomorphic processes occurring at the micro-scale and at short timescale (e.g. development of individual plant response traits to mechanical stress) on those occurring at larger spatio-temporal scales (e.g. construction of wooded pioneer fluvial islands, benches and floodplains; adjustment of the landscape mosaic). We also stress the need to further develop nested hierarchical biogeomorphic bottom-up and top-down cross scale studies within fluvial corridors to better understand complex biogeomorphic fluvial ecosystems.

Haut de page

Bibliographie

Abernethy B., Rutherfurd I. D. (1998) – Where along a river’s length will vegetation most effectively stabilise stream banks? Geomorphology, 23 (1), 55-75.
DOI : 10.1016/S0169-555X(97)00089-5

Bornette G., Tabacchi E., Hupp C., Puijalon S., Rostan J.C. (2008) – A model of plant strategies in fluvial hydrosystems. Freshwater Biology, 53 (8), 1692-1705.
DOI : 10.1111/j.1365-2427.2008.01994.x

Bouma T.J., Temmerman S., van Duren L.A., Martini E., Vandenbruwaene W., Callaghan D.P., Balke T., Biermans G., Klaassen P.C., van Steeg P., Dekker F., van de Koppel J., Vries M.B. de, Herman. P.M.J. (2013) – Organism traits determine the strength of scale-dependent bio-geomorphic feedbacks: A flume study on three intertidal plant species. Geomorphology, 180-181, 57-65.
DOI : 10.1016/j.geomorph.2012.09.005

Chamaillard S. (2011) – Efficience d’utilisation de l’eau chez le peuplier noir (Populus nigra L.): variabilité et plasticité en réponse aux variations de l’environnement. PhD thesis, Université d'Orléans, 188 p.

Cooper D.J., Merritt D.M., Andersen D.C., Chimner R.A. (1999) – Factors controlling the establishment of Fremont cottonwood seedlings on the Upper Green River, USA. Regulated Rivers: Research & Management, 15 (5), 419-440.
DOI : 10.1002/(SICI)1099-1646(199909/10)15:5<419::AID-RRR555>3.0.CO;2-Y

Corenblit D., Baas A., Balke T., Bouma T., Fromard F., Garófano-Gómez V., González E., Gurnell A.M., Hortobágyi B., Julien F., Kim D., Lambs L., Stallins J.A., Steiger J., Tabacchi E., Walcker R. (2015) – Engineer pioneer plants respond to and affect geomorphic constraints similarly along water-terrestrial interfaces world-wide: Biogeomorphic feedbacks along water-terrestrial interfaces. Global Ecology and Biogeography, 24 (12), 1363-1376.
DOI : 10.1111/geb.12373

Corenblit D., Steiger J., Charrier G., Darrozes J., Garófano-Gómez V., Garreau A., González E., Gurnell A. M., Hortobágyi B., Julien F., Lambs L., Larrue S., Otto T., Roussel E., Vautier F., Voldoire O. (2016) – Populus nigra L. establishment and fluvial landform construction: biogeomorphic dynamics within a channelized river: Biogeomorphic dynamics within a channelized river. Earth Surface Processes and Landforms, 41 (9), 1276-1292.
DOI : 10.1002/esp.3954

Corenblit D., Steiger J., González E., Gurnell A.M., Charrier G., Darrozes J., Dousseau J., Julien F., Lambs L., Larrue S., Roussel E., Vautier F., Voldoire O. (2014) – The biogeomorphological life cycle of poplars during the fluvial biogeomorphological succession: a special focus on Populus nigra L. Earth Surface Processes and Landforms, 39 (4), 546-563.
DOI : 10.1002/esp.3515

Corenblit D., Tabacchi E., Steiger J., Gurnell A.M. (2007) – Reciprocal interactions and adjustments between fluvial landforms and vegetation dynamics in river corridors: A review of complementary approaches. Earth-Science Reviews, 84 (1-2), 56-86.
DOI : 10.1016/j.earscirev.2007.05.004

Dech J.P., Maun M.A. (2006) – Adventitious root production and plastic resource allocation to biomass determine burial tolerance in woody plants from central Canadian coastal dunes. Annals of Botany, 98 (5), 1095-1105.
DOI : 10.1093/aob/mcl196

Dejaifve P.A., Esquirol N. (2011) – Quantification et répartition du bois mort échoué dans la Réserve Naturelle Nationale du Val d’Allier. Moulins: Réserve Naturelle Nationale du Val d’Allier - DIREN-Auvergne, 24 p.

Diehl R.M., Wilcox A.C., Stella J.C., Kui L., Sklar L.S., Lightbody A. (in press) – Fluvial sediment supply and pioneer woody seedlings as a control on bar-surface topography. Earth Surface Processes and Landforms.
DOI : 10.1002/esp.4017

Ding Z. (2014) – Réponse du système racinaire de Populus nigra L. aux contraintes hydrogéomorphologiques sur les bancs alluviaux de l'Allier. Master thesis. Université Blaise Pascal, 57 p.

Eichel J., Corenblit D., Dikau R. (2015) – Conditions for feedbacks between geomorphic and vegetation dynamics on lateral moraine slopes: a biogeomorphic feedback window. Earth Surface Processes and Landforms, 41, 406-419.
DOI:10.1002/esp.3859

Ennos A.R. (1993) – The scaling of root anchorage. Journal of Theoretical Biology, 161 (1), 61-75.
DOI : 10.1006/jtbi.1993.1040

Erktan A., Rey F. (2013) – Linking sediment trapping efficiency with morphological traits of Salix tiller barriers on marly gully floors under ecological rehabilitation. Ecological Engineering, 51, 212-220.
DOI : 10.1016/j.ecoleng.2012.12.003

Euler T., Zemke J., Rodrigues S., Herget J. (2014) – Influence of inclination and permeability of solitary woody riparian plants on local hydraulic and sedimentary processes. Hydrological Processes, 28 (3), 1358-1371.
DOI :10.1002/hyp.9655

Francis R.A., Corenblit D., Edwards P.J. (2009) – Perspectives on biogeomorphology, ecosystem engineering and self-organisation in island-braided fluvial ecosystems. Aquatic Sciences, 71 (3), 290-304.
DOI : 10.1007/s00027-009-9182-6

Goodman A. M., Ennos A.R. (1996) – A comparative study of the response of the roots and shoots of sunflower and maize to mechanical stimulation. Journal of Experimental Botany, 47 (10), 1499-1507.
DOI : 10.1093/jxb/47.10.1499

Gurnell A. (2014) – Plants as river system engineers. Earth Surface Processes and Landforms, 39 (1), 4-25.
DOI : 10.1002/esp.3397

Hortobágyi B., Corenblit D., Vautier F., Steiger J., Roussel E., Burkart A., Peiry J.L. (in press) – A multi-scale approach of fluvial biogeomorphic dynamics using photogrammetry. Journal of Environmental Management.
DOI : 10.1016/j.jenvman.2016.08.069

Hortobágyi B., Corenblit D., Steiger J., Peiry J.L. (accepted) – Niche construction within riparian corridors. Part I: Exploring biogeomorphic feedback windows of three pioneer riparian species, Allier River, France. Geomorphology

Jones C.G., Lawton J.H., Shachak M. (1994) – Organisms as ecosystem engineers. Oikos, 69, 373-386.
DOI : 10.2307/3545850

Karrenberg S., Blaser S., Kollmann J., Speck T., Edwards P.J. (2003) – Root anchorage of saplings and cuttings of woody pioneer species in a riparian environment. Functional Ecology, 17 (2), 170-177.
DOI : 10.1046/j.1365-2435.2003.00709.x

Karrenberg S., Edwards P.J., Kollmann J. (2002) – The life history of Salicaceae living in the active zone of floodplains. Freshwater Biology, 47 (4), 733-748.
DOI : 10.1046/j.1365-2427.2002.00894.x

Kui L., Stella J.C. (2016) – Fluvial sediment burial increases mortality of young riparian trees but induces compensatory growth response in survivors. Forest Ecology and Management, 366, 32-40.
DOI : 10.1016/j.foreco.2016.02.001

Kui L., Stella J.C., Lightbody A., Wilcox A.C. (2014) – Ecogeomorphic feedbacks and flood loss of riparian tree seedlings in meandering channel experiments. Water Resources Research, 50 (12), 9366-9384.
DOI : 10.1002/2014WR015719

Manners R.B., Wilcox A.C., Kui L., Lightbody A.F., Stella J.C., Sklar L.S. (2015) – When do plants modify fluvial processes? Plant-hydraulic interactions under variable flow and sediment supply rates. Journal of Geophysical Research: Earth Surface, 120 (2), 325-345.
DOI : 10.1002/2014JF003265

Moggridge H.L., Gurnell A.M. (2009) – Controls on the sexual and asexual regeneration of Salicaceae along a highly dynamic, braided river system. Aquatic Sciences, 71 (3), 305-317.
DOI : 10.1007/s00027-009-9193-3

Pasquale N., Perona P., Francis R., Burlando P. (2013) – Above-ground and below-ground Salix dynamics in response to river processes. Hydrological Processes, 28 (20), 5189-5203.
DOI : 10.1002/hyp.9993

Petit S. (2006) – Reconstitution de la dynamique du paysage alluvial de trois secteurs fonctionnels de la rivière Allier (1946-2000), Massif Central, France. Géographie physique et Quaternaire, 60 (3), 271-287.
DOI : 10.7202/018000ar

Polvi L.E., Wohl E., Merritt D.M. (2014) – Modeling the functional influence of vegetation type on streambank cohesion. Earth Surface Processes and Landforms, 39 (9), 1245-1258.
DOI : 10.1002/esp.3577

Puijalon S., Bornette G. (2004) – Morphological variation of two taxonomically distant plant species along a natural flow velocity gradient. New Phytologist, 163 (3), 651-660.
DOI : 10.1111/j.1469-8137.2004.01135.x

Puijalon S., Bouma T.J., Douady C.J., van Groenendael J., Anten N.P.R., Martel E., Bornette G. (2011) – Plant resistance to mechanical stress: evidence of an avoidance-tolerance trade-off. New Phytologist, 191 (4), 1141-1149.
DOI : 10.1111/j.1469-8137.2011.03763.x

Puijalon S., Léna J.P., Rivière N., Champagne J.Y., Rostan J.C., Bornette G. (2008) – Phenotypic plasticity in response to mechanical stress: hydrodynamic performance and fitness of four aquatic plant species. New Phytologist, 177 (4), 907-917.
DOI : 10.1111/j.1469-8137.2007.02314.x

Read J., Stokes A. (2006) – Plant biomechanics in an ecological context. American journal of botany, 93 (10), 1546-1565.
DOI : 10.3732/ajb.93.10.1546

Rivière A. (1977) – Méthodes granulométriques, techniques et interprétation. Masson. Paris, 170 p.

Steiger J., Gurnell A.M. (2003) – Spatial hydrogeomorphological influences on sediment and nutrient deposition in riparian zones: observations from the Garonne River, France. Geomorphology, 49 (1-2), 1-23.
DOI : 10.1016/S0169-555X(02)00144-7

Stokes A., Fitter A.H., Courts M.P. (1995) – Responses of young trees to wind and shading: effects on root architecture. Journal of Experimental Botany, 46 (9), 1139-1146.
DOI : 10.1093/jxb/46.9.1139

Tamasi E., Stokes A., Lasserre B., Danjon F., Berthier S., Fourcaud T., Chiatante D. (2005) – Influence of wind loading on root system development and architecture in oak (Quercus robur L.) seedlings. Trees, 19 (4), 374-384.
DOI : 10.1007/s00468-004-0396-x

Telewski F.W., Pruyn M.L. (1998) – Thigmomorphogenesis: a dose response to flexing in Ulmus americana seedlings. Tree physiology, 18 (1), 65-68.
DOI : 10.1093/treephys/18.1.65

Violle C., Navas M. L., Vile D., Kazakou E., Fortunel C., Hummel I., Garnier E. (2007) – Let the concept of trait be functional! Oikos, 116 (5), 882-892.
DOI : 10.1111/j.0030-1299.2007.15559.x

Haut de page

Annexe

Version française abrégée

Les arbres riverains pionniers, tels que Populus nigra L., établis sur des bancs alluviaux au sein de corridors fluviaux dynamiques, influencent fortement la géomorphologie fluviale en piégeant des sédiments et en construisant des formes fluviales pendant les crues (Corenblit et al., 2007 ; Gurnell, 2014 ; Hortobágyi et al., in press). L'effet ingénieur (changements dans l’état physique de l’habitat par les organismes) de P. nigra sur les bancs alluviaux dépend de sa biomasse et de son degré d’exposition à la contrainte mécanique. P. nigra a une plasticité phénotypique élevée qui sous-tend une grande variation de ses traits morphologiques et biomécaniques selon les conditions hydrogéomorphologiques locales. La compréhension et la quantification de la variation des traits de réponses morphologiques et biomécaniques au sein des populations de P. nigra en fonction de leur exposition à la contrainte mécanique est fondamentale pour mieux comprendre pourquoi et comment les plantes sont capables d'influer sur la géomorphologie. Dans une étude empirique in situ, nous avons quantifié la variation des traits de réponses morphologiques et biomécaniques de populations de P. nigra selon trois niveaux différents de contrainte mécanique. En contexte très exposé (en tête de banc alluvial), les plantes ont développé des traits de réponses permettant une plus grande résistance, notamment une taille réduite, une tige flexible et inclinée et un système racinaire plus robuste (tab. 1, fig. 6). Néanmoins, ces réponses morphologiques et biomécaniques réduisent en même temps le potentiel « ingénieur » des plantes, c’est-à-dire la capacité de piégeage des sédiments fins. Hortobágyi et al. (accepted) ont démontré sur la rivière Allier l’effet ingénieur limité du peuplier noir dans les zones les plus exposées, ce qui traduit un effet inhibiteur de la construction de niche par les contraintes mécaniques liées à l’écoulement. En contextes moins exposés (en queue de banc et dans le chenal secondaire), les peupliers ont développé une tige plus large, plus longue, moins flexible et moins inclinée. Cette réponse exprime une fonction préférentiellement dédiée à l’acquisition des ressources et à la bioconstruction. Le fait que la longueur totale des plantes est maximale en queue de banc suggère que l’enfouissement par des sédiments fins stimule la production de la biomasse aérienne, celle-ci renforçant le potentiel photosynthétique. La construction des formes fluviales par le peuplier noir peut être considérée dans notre cas d’étude et de manière plus générale comme une stratégie de construction de niche conduisant à la diminution progressive des contraintes mécaniques et à l’accumulation des sédiments fins, de la matière organique et des nutriments qui influencent de manière positive la survie et la croissance des peupliers (Corenblit et al., 2014). Nos résultats suggèrent que les processus biogéomorphologiques se produisant à une échelle fine ont une influence sur les processus à une échelle plus large. Cela pointe la nécessité de développer des études biogéomorphologiques hiérarchiques selon des approches ascendantes (bottom-up) et descendante (top-down) pour renforcer notre compréhension du fonctionnement des écosystèmes biogéomorphologiques fluviaux.

Haut de page

Table des illustrations

Titre Fig. 1 – Localisation map. Fig. 1 Carte de localisation.
Légende A. Study site; B. Sampling sites on the Allier River. C. Bar-head; D. Bar-tail; E. Lower reach of chute channel A. Zone d’étude ; B. Sites d’échantillonnage sur la rivière Allier. C. Tête de banc ; D. Queue de banc ; E. Tronçon aval de la chute alluviale.
URL http://geomorphologie.revues.org/docannexe/image/11748/img-1.png
Fichier image/png, 11M
Titre Fig. 2 – Daily mean discharges (m3.s-1) at the hydrological station of Châtel-de-Neuvre between 2011 and 2014 and possible recruitment periods of P. nigra populations (data. Banque Hydro http://www.hydro.eaufrance.fr; station code: K3400810).Fig. 2 – Débits journaliers (m3/s) à la station hydrologique de Châtel-de-Neuvre entre 2011 et 2014 et les périodes de recrutement possible des populations de P. nigra (données. Banque Hydro http://www.hydro.eaufrance.fr ; code de station : K3400810).
URL http://geomorphologie.revues.org/docannexe/image/11748/img-2.png
Fichier image/png, 526k
Titre Fig. 3 – Surface and subsurface cumulated particle size distributions of the three sample sites.Fig. 3 – Courbes granulométriques cumulées des trois sites à la surface et à la subsurface.
URL http://geomorphologie.revues.org/docannexe/image/11748/img-3.png
Fichier image/png, 199k
Titre Tab. 1 – Summary statistics and results of Kruskal-Wallis test (α = 0.05) of aboveground and belowground P. nigra traits. Letters assigned to groups (i.e. A, B) are indicating that the groups are significantly different. Tab. 1 Statistiques sommaires et résultats du test de Kruskal-Wallis (α = 0,05) des traits aériens et souterrains du P. nigra. Les lettres associées aux groupes (i.e. A, B) indiquent que les groupes sont significativement différents.
URL http://geomorphologie.revues.org/docannexe/image/11748/img-4.png
Fichier image/png, 225k
Titre Fig. 4 – Discriminant analysis based on above- and belowground response traits.Fig. 4 Analyse factorielle discriminante basée sur des traits aériens et souterrains.
Légende A. Two-dimensional chart representing the observations on the factor axes; B. Correlation between the initial variables and the two factors. 1. Bar-head; 2. Bar‑tail; 3. Lower reach of chute channel. A. Diagramme bi-dimensionnel représentant les observations sur les axes factoriels ; B. Corrélation entre les variables initiales et les deux facteurs. 1. Tête de banc ; 2. Queue de banc ; 3. Tronçon aval de la chute alluviale.
URL http://geomorphologie.revues.org/docannexe/image/11748/img-5.png
Fichier image/png, 405k
Titre Tab. 2 – Confusion matrix for the estimation sample and for the cross-validation.Tab. 2 – Matrices de confusion pour l’échantillon d’estimation et pour les résultats de la validation croisée.
URL http://geomorphologie.revues.org/docannexe/image/11748/img-6.png
Fichier image/png, 45k
Titre Fig. 5 – Membership probabilities and membership classes of observations deduced by the probabilities in cross-validation mode. Fig. 5 Probabilité d’appartenance à un groupe et classes d’appartenance des observations déduites par les probabilités en mode de validation croisée.
Légende A. Membership probability to be assigned to the bar-head population; B. Membership probability to be assigned to the bar-tail population; C. Membership probability to be assigned to the chute channel population. Note that thresholds between classes are different for A, B and C (see legend). A. Probabilité d’appartenance attribuée à la tête de banc ; B. Probabilité d’appartenance attribué à la queue de banc ; C. Probabilité d’appartenance à être attribuée à la chute alluviale. Notez que les seuils de classes sont différents pour chaque site (voir légende).
URL http://geomorphologie.revues.org/docannexe/image/11748/img-7.png
Fichier image/png, 2,8M
Titre Fig. 6 – Schematic representation of Populus nigra L. exposed to different levels of mechanical stress. The length of adventitious and lateral roots remains unknown since they could not be excavated entirely. Question marks indicate that root lengths are unknown.Fig. 6 Représentation schématique des Populus nigra L. exposés au stress mécanique de différents niveaux. La longueur des racines adventives et latérales est inconnue, étant donné qu’elles n’ont pas été excavées dans leur intégralité. Les points d’interrogation indiquent le manque d’information sur la longueur des racines.
URL http://geomorphologie.revues.org/docannexe/image/11748/img-8.png
Fichier image/png, 2,5M
Haut de page

Pour citer cet article

Référence électronique

Borbála Hortobágyi, Dov Corenblit, ZhuQing Ding, Luc Lambs et Johannes Steiger, « Above- and belowground responses of Populus nigra L. to mechanical stress observed on the Allier River, France », Géomorphologie : relief, processus, environnement [En ligne], Articles sous presse, mis en ligne le 11 septembre 2017, consulté le 20 septembre 2017. URL : http://geomorphologie.revues.org/11748

Haut de page

Auteurs

Borbála Hortobágyi

Université Clermont Auvergne, CNRS-GEOLAB – 63000 Clermont-Ferrand, France | LTER France, LTSER Zone Atelier Loire – 37000 Tours, France (borbala.hortobagyi@uca.fr). Tel : +33 4 73 34 68 23 / Fax : +33 4 73 34 68 24

Articles du même auteur

Dov Corenblit

Université Clermont Auvergne, CNRS-GEOLAB – 63000 Clermont-Ferrand, France | LTER France, LTSER Zone Atelier Loire –37000 Tours, France (dov.corenblit@uca.fr).

Articles du même auteur

ZhuQing Ding

Université Clermont Auvergne, CNRS-GEOLAB – 63000 Clermont-Ferrand, France (dzqdzq1988@163.com).

Luc Lambs

Université Paul Sabatier, INP, CNRS-UMR 5245, ECOLAB-Laboratoire écologie fonctionnelle et environnement, Université Toulouse III –31062 Toulouse, France (luc.lambs@univ-tlse3.fr).

Johannes Steiger

Université Clermont Auvergne, CNRS-GEOLAB – F-63000 Clermont-Ferrand, France | LTER France, LTSER Zone Atelier Loire – 37000 Tours, France (johannes.steiger@uca.fr).

Articles du même auteur

Haut de page

Droits d’auteur

© Groupe français de géomorphologie

Haut de page
  • Logo CNRS - Institut des sciences humaines et sociales
  • Logo Groupe français de géomorphologie
  • Revues.org