Differences in flooding tolerance between species from two wetland habitats with contrasting hydrology: implications for vegetation development in future floodwater retention areas

Differences in flooding tolerance between species from two wetland habitats withcontrasting hydrology: implications for vegetation development in future

floodwater retention areas

Katarzyna Banach1,2, Artur M. Banach1,2, Leon P. M. Lamers2, Hans De Kroon2, Riccardo P. Bennicelli1,Antoine J. M. Smits3 and Eric J. W. Visser2,*

1Department of Biochemistry and Environmental Chemistry, The John Paul II Catholic University of Lublin, Al. Krasnicka 102,20-718 Lublin, Poland, 2Department of Ecology, Institute for Water and Wetland Research, Radboud University Nijmegen,

Heyendaalseweg 135, 6525 AJ Nijmegen, The Netherlands and 3Centre for Sustainable Management of Resources, RadboudUniversity Nijmegen, Heyendaalseweg 135, 6525 AJ Nijmegen, The Netherlands

Received: 28 February 2008 Returned for revision: 8 June 2008 Accepted: 11 August 2008 Published electronically: 3 October 2008

† Background and Aims Plants need different survival strategies in habitats differing in hydrological regimes.This probably has consequences for vegetation development when former floodplain areas that are currently con-fronted with soil flooding only, will be reconnected to the highly dynamical river bed. Such changes in river man-agement are increasingly important, especially at locations where increased water retention can prevent floodingevents in developed areas. It is therefore crucial to determine the responses of plant species from relatively low-dynamic wetlands to complete submergence, and to compare these with those of species from river forelands, inorder to find out what the effects of such landscape-scale changes on vegetation would be.† Methods To compare the species’ tolerance to complete submergence and their acclimation patterns, a green-house experiment was designed with a selection of 19 species from two contrasting sites: permanently wetmeadows in a former river foreland, and frequently submerged grasslands in a current river foreland. Theplants were treated with short (3 weeks) and long (6 weeks) periods of complete submergence, to evaluate ifsurvival, morphological responses, and changes in biomass differed between species of the two habitats.† Key Results All tested species inhabiting river forelands were classified as tolerant to complete submergence,whereas species from wet meadows showed either relatively intolerant, intermediate or tolerant responses.Species from floodplains showed in all treatments stronger shoot elongation, as well as higher production ofbiomass of leaves, stems, fine roots and taproots, compared with meadow species.† Conclusions There is a strong need for the creation of temporary water retention basins during high levels ofriver discharge. However, based on the data presented, it is concluded that such reconnection of former wetlands(currently serving as meadows) to the main river bed will strongly influence plant species composition andabundance.

Key words: Acclimation, biomass allocation, climate, complete submergence, flooding tolerance, retentionareas, shoot elongation, soil flooding, waterlogging, wetland species.

INTRODUCTION

Flooding of riverine areas can be an important stress factor forplants, when these cannot acclimate to the adverse conditionsduring submergence. Such flooding events have always beentemporal, as they are caused by strong rainfall and snowmelting, but recently peak discharges of European rivershave increased in volume due to extensive anthropogenicactivities (such as regulation of rivers, removal of vegetationcover over large areas, intensive agriculture and building ofdams and roads) (Blom and Voesenek, 1996). Human develop-ment is often near river beds, which is why it becomes moreand more important to control flooding, in order to preventsubstantial personal and financial damage. Reconnection offormer river floodplains has been proposed as a measure towiden the river bed in periods of high water discharge, butso far it is not known how vegetation in these areas willrespond to such a dramatic change in flooding regime. This

study compares the flooding tolerance of plant speciesadapted to frequent river floods with that of species fromformer river forelands, thus yielding information on whichchanges in vegetation may be expected if such areas are recon-nected to the current river bed.

Soil flooding causes displacement of gases when soil poresare filled with water. The low diffusion rate of oxygen in thismedium (which is 10 000 times slower than in air; Armstrong,1979) results in limitation of oxygen availability for plantroots, soil micro-organisms and chemical processes (Glinskiand Stepniewski, 1985), and leads to a switch of aerobicmetabolism of plants into less efficient anaerobic fermentation,causing a fast depletion of carbohydrate reserves(Bailey-Serres and Voesenek, 2008) and, in intolerantspecies, ultimately in plant death (Fox et al., 1995; Gibbsand Greenway, 2003). Additionally, accumulation in the soilof reduced phytotoxins (Fe2þ, Mn2þ, sulfide and, at high con-centrations, ammonium; Snowden and Wheeler, 1993;Lucassen et al., 2000, 2002) can have a negative impact on

* For correspondence. E-mail [email protected]

# The Author 2008. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.

For Permissions, please email: [email protected]

Annals of Botany 103: 341–351, 2009

doi:10.1093/aob/mcn183, available online at www.aob.oxfordjournals.org

plants, causing, among others, growth retardation, reduction inleaf size, wilting of shoots and necrosis (Snowden andWheeler, 1993).

In response to the severity of flooding stress, terrestrialwetland species developed a variety of strategies to resistflooding (Vartapetian and Jackson, 1997; Bailey-Serres andVoesenek, 2008). High metabolic activity may be avoidedduring submergence in order to lower energy demand(Geigenberger, 2003), thereby saving carbohydrate reserves.Alternatively, increased shoot elongation can restore contactof leaves with the atmosphere (Banga et al., 1995; Voeseneket al., 2004), which combined with improved internal gastransport via aerenchyma formation results in a sufficientlyhigh oxygen status (Armstrong, 1972; Visser et al., 1996;McDonald et al., 2002; Colmer, 2003). If the water surfacecannot be reached, underwater photosynthesis may improvecarbohydrate and oxygen concentrations (Laan and Blom,1990; Mommer and Visser, 2005). Ultimately, survivalduring submergence will depend on the balance betweencosts and benefits of these acclimations, which are largelydetermined by timing, frequency, depth and duration of flood-ing (van Eck et al., 2004, 2005; Voesenek et al., 2004), andalso by the interaction between the floodwater properties andbiochemical processes in the soil (van der Welle et al., 2007).

The flooding tolerance of species determines their distri-bution along flooding gradients in river forelands (Blomet al., 1990; van Eck et al., 2004). However, it is obviousthat tolerance to soil flooding (waterlogging) requires a differ-ent suite of traits than tolerance to complete submergencedoes. Traits that improve internal gas transport betweenshoot and roots (Colmer, 2003) and that decrease vulnerabilityto toxic reduced soil components (Laan et al., 1989; Engelaaret al., 1995; Colmer, 2003) are imperative during soil flooding,whereas surviving complete submergence requires additionalresponses such as shoot elongation, underwater photosynthesisand/or a much more efficient use of carbohydrate stores(Mommer and Visser, 2005). This does not exclude that asingle species can display both types of traits, as shown byRumex crispus, which forms adventitious, aerenchymatousroots within just a few days after soil flooding (Visser et al.,1996), but also survives complete submergence under lowlight conditions for .2 years (Vervuren et al., 2003).However, adaptation to both types of flooding regimes is prob-ably not very common, as indicated by the differences inspecies composition between habitats with permanent soilflooding and those with frequent deep flooding. Still, exper-imental screening of differences in tolerance to complete sub-mergence of species from such contrasting wetland habitatshas not been done so far.

If different strategies are needed in habitats differing inflooding regime, then this will have consequences for plantperformance when hydrological conditions are altered (vanBodegom et al., 2006), for instance when areas that havebeen confronted for a long time with regular soil floodingonly, are reconnected to the highly dynamical river bed.Such changes in river management are increasingly urgent(Wolsink, 2006), especially at locations where water retentioncan prevent flooding events in important urban, industrial andagricultural areas. The frequency of summer flooding eventshas increased during the last decades, because of emissions

of greenhouse gases enhancing global warming and concomi-tantly hydrological cycles (Ulbricht et al., 2003; Kundzewiczet al., 2005). It is therefore crucial to know the responses ofspecies from relatively low-dynamic wetlands to deep flood-ing, and compare these with those of species from river fore-lands, in order to find out what the effects of suchlandscape-scale changes on vegetation would be.

In order to compare tolerance of species to flooding andtheir acclimation patterns, a greenhouse experiment wasdesigned with 19 plant species from two contrasting sites:(1) permanently wet meadows in a former river foreland; (2)frequently submerged grasslands in a river foreland.Treatments with short (3 weeks) and longer (6 weeks)periods of complete submergence were used to find out ifthese species differed in their survival, morphologicalresponses and growth upon deep flooding.

MATERIALS AND METHODS

Research area

Plant material was collected from grasslands near the villagesKepa Solecka (518080N 218470E) and Kosiorow (518130N218530E), which are located in south-east Poland. Theseresearch areas differ in hydrological conditions and vegetationcomposition, as described below.

Seeds from Kepa Solecka were collected from grasslands onsandy clay in the river forelands of the Vistula River. This areais irregularly but frequently flooded in periods of winter orspring snow melting and after strong spring, summer andautumn rainfall. The vegetation distribution is influenced bythese flooding events, as can be seen from a zonation thatdepends on the elevational gradient and on the distance fromthe river bed (data not shown). Plant material from Kosiorowwas collected from flat, very moist meadows with peaty soil.In the past, this area was regularly flooded by the ChodelkaRiver, a tributary of Vistula River, but dike building has pro-tected it against flooding for 50 years. The current vegetationis dominated by Deschampsia cespitosa and Holcus lanatus,and the vegetation composition varies with soil moisture,content of soil organic matter and agricultural management(e.g. mowing, grazing, fertilization), but many species seemto be generally well adapted to high soil water tables.

Plant material

Nineteen plant species, including eight species from theKepa Solecka floodplain and 11 species from the Kosiorowmeadows, were selected based on abundance in the field,seed availability and germination success. According toEllenberg’s indicator values (Ellenberg et al., 1992), thesespecies represent a range of preferences to different soil moist-ure conditions, although on average the species tended toprefer relatively moist to wet soil (5–7 on the scale ofEllenberg; Table 1). It should be noted that this scale doesnot discriminate between habitats with fluctuating watertables and stable/stagnant soil water.

The experiment was performed in two series, conducted inautumn and winter of two consecutive years, becauseof limitations in seed/ramet availability and space. These

Banach et al. — Flooding tolerance in two contrasting wetland habitats342

two autumn/winter periods did not show extreme warmperiods, which made control of greenhouse conditions (seebelow) very much comparable between the two years. Themajority of plants (Achillea millefolium, Arabidopsissuecica, Cerastium fontanum, Deschampsia cespitosa,Linaria vulgaris, Lolium perenne, Phleum pratense,Plantago lanceolata, Plantago major, Prunella vulgaris,Ranunculus acris, Rumex acetosa, Rumex confertus, Rumexcrispus, Silene pratensis and Verbascum densiflorum) wereraised from seeds collected in one of the research areas. Theseeds were germinated on moist filter paper in Petri dishes ina growth cabinet [photosynthetic photon flux density (PPFD)20 mmol m22 s21 at plant level, day 12 h 20 8C, night 12 h10 8C] for about 1 week. Then, the seedlings were transplantedonto polyethylene granules soaked with nutrient solution(Visser et al., 1996) and placed in a climate room (PPFD200 mmol m22 s21 during 16 h per day, temperature 22 8C,air humidity 50 %) for 3 weeks. Galium boreale, Galiumpalustre and Holcus lanatus plants originated from cuttingsmade from plants collected from meadows in Kosiorow,since insufficient viable seeds could be collected. Theseplants were pre-grown in the greenhouse (light conditions.100 mmol m22 s21 PPFD by additional lighting for 16 hper day, temperature approx. 20 8C).

After 3 weeks all plant species were planted into polyethy-lene pots (8 � 8 � 9 cm) filled with a mixture (1 : 1, v/v) ofsieved fertilized potting soil and river sand, and placed inthe greenhouse for 2 weeks to acclimatize. The positions ofthe pots were changed regularly to avoid the influence oflocal conditions on growth of replicate plants.

Experimental design

Well-developed young plants were selected for the exper-iment, distributing size classes equally over the treatments.An initial group was harvested at the beginning of the

experiment in order to determine initial size and weight ofthe plants. Plants of each species were subjected to four treat-ments with flooding and recovery periods as shown in Fig. 1.Control plants were grown for 8 weeks with watering threetimes a week. Two groups received a short flooding of 3weeks, followed by either 2 weeks or 5 weeks of recoveryunder control conditions. Finally, one group received a longflooding of 6 weeks followed by 2 weeks of recovery.

The durations of the flooding periods were chosen accordingto data from previous experiments (Blom et al., 1990; van Ecket al., 2004), where these durations proved to result in substan-tially different responses in tolerant and intolerant species. Arecovery period in the experimental design is essential to dis-tinguish between dead and viable individuals; immediatelyafter de-submergence this is not feasible. Furthermore,re-aeration after flooding can lead to the formation of freeoxygen radicals and post-anoxic injury (Crawford andBrandle, 1996). The damage in plant cells caused by the rad-icals can result in death of plants within 2 weeks of flooding, ifthese do not possess mechanisms preventing such excessiveoxidation processes (Nabben et al., 1999). Two different dur-ations of recovery following short flooding were chosen to beable to compare the effect of short flooding with control plantgrowth (no flooding for 8 weeks), and, alternatively, tocompare two different flooding durations (short and longflooding) followed by a similar recovery period.

Plants subjected to flooding treatments were randomlyplaced in two basins (diameter 1.85 m, depth 0.9 m) in thegreenhouse, which were filled with tap water 2 weeks beforethe beginning of the experiment. The depth of water columnwas kept to 60 cm above pot level in order to ensure completesubmergence of all plants. The water was mechanically circu-lated and filtered to reduce algae growth, and the filtered waterreturned to the basin via a small waterfall, thereby restoring theequilibrium of the dissolved gases with the air. Dissolvedcarbon dioxide concentrations were around 15 mM (at 20 8C

TABLE 1. Plant species used in the experiment, including a brief description of their habitat and indicator values for soil moisturebased on Ellenberg et al. (1992)

Species Abbreviation Family Habitat Ellenberg’s value Origin

Achillea millefolium Am Asteraceae Dry 4 River forelandArabidopsis suecica As Brassicaceae * – Wet meadowCerastium fontanum Cf Caryophyllaceae Intermediate moist 5 Wet meadowDeschampsia cespitosa Dc Poaceae Very moist 7 Wet meadowGalium boreale Gb Rubiaceae Intermediate to very moist 6 Wet meadowGalium palustre Gp Rubiaceae Wet, oxygen poor soil 9 Wet meadowHolcus lanatus Hl Poaceae Intermediate to very moist 6 Wet meadowLinaria vulgaris Lv Scrophulariaceae Dry 4 Wet meadowLolium perenne Lp Poaceae Intermediate moist 5 River forelandPhleum pratense Pp Poaceae Intermediate moist 5 River forelandPlantago lanceolata Pl Plantaginaceae Indifferent x River forelandPlantago major Pm Plantaginaceae Intermediate moist 5 River forelandPrunella vulgaris Pv Lamiaceae Intermediate moist 5 River forelandRanunculus acris Racr Ranunculaceae Intermediate to very moist 6 Wet meadowRumex acetosa Ra Polygonaceae Indifferent – Wet meadowRumex confertus Rco Polygonaceae * – River forelandRumex crispus Rcr Polygonaceae Very moist 7 River forelandSilene pratensis Sp Caryophyllaceae Dry 4 Wet meadowVerbascum densiflorum Vd Scrophulariaceae Dry 4 Wet meadow

*, Not listed in Ellenberg et al. (1992).

Banach et al. — Flooding tolerance in two contrasting wetland habitats 343

and pH 8.4), as measured earlier in similar experiments(Mommer et al., 2006).

The PPFD was regularly measured in basins and ranged atplant level between 7 and 24 mmol m22 s21. After 5 and 8weeks, depending on the treatments, survival of plants wasdetermined based on plant appearance. Plants with any greenturgid leaves were assigned as viable (Nabben et al., 1999;van Eck et al., 2004). When .50 % of the plants of a speciesdied after the short flooding treatment of 3 weeks, this specieswas supposed to be relatively intolerant. When .50 % of theplants died after the long flooding treatment, a species wasclassified as intermediately tolerant, and when .50 % of theplants survived 6 weeks of flooding it was rated as tolerant,similar to the classification as proposed by Blom et al. (1990)and van Eck et al. (2004).

At harvest, height of the shoot and length of the longest leafwere measured of the surviving plants in each treatment, inorder to estimate which flooding depth could be overgrownby the shoot, after which the root system was gently washedout of the soil. No clear indications of excessive release oftoxic compounds from the soil in the submerged pots, suchas precipitations of Fe or sulphide, were found. Plants weredivided into leaves (including petioles), stems, taproots andfine roots and dried at 70 8C for 48 h.

Data analysis

Statistical testing of data on dry weight of leaves, stems, tap-roots, fine roots and plant height did not include relativelyintolerant species, because in these cases the number of repli-cates was insufficient. Data were log (x þ 1) transformed tomeet the assumptions of normal distribution and homogeneityof variance, and analysed with SPSS version 15.0 (SPSS Inc.,Chicago, IL, USA) and SAS version 9.1 (SAS Institute Inc.,Cary, NC, USA).

The data were examined with two two-way nestedANOVAs, with treatment and habitat or tolerance level asfixed factors and species as random factor. In the firstanalysis, species were nested within habitat, whereas in the

second analysis species were nested within tolerance level.Treatment means were compared with Tukey’s tests.

RESULTS

Survival

The survival during flooding treatments was significantlydifferent (P , 0.001) for plant species from the two researchareas (Fig. 2A). All species inhabiting river foreland survivedboth 3 and 6 weeks of flooding and were thus classified as tol-erant (for classification criteria, see Materials and methods),whereas species from wet meadows showed one of threeresponses: relatively intolerant (three species – Galium palus-tre, Silene pratensis and Verbascum densiflorum), intermediate(six species – Arabidopsis suecica, Cerastium fontanum,Galium boreale, Holcus lanatus, Linaria vulgaris and Rumexacetosa) or tolerant (two species – Deschampsia cespitosaand Ranunculus acris).

Among the species that were tolerant to flooding, eight ofthem (Deschampsia cespitosa, Plantago lanceolata, Plantagomajor, Phleum pratense, Prunella vulgaris, Rumex acetosa,Rumex confertus and Rumex crispus) showed no mortality atall, whereas survival rates of Achillea millefolium, Loliumperenne and Ranunculus acris decreased by 20–30 % after 6weeks of flooding (Fig. 2A).

In order to allow comparisons between plants with the samegrowth period above water, and also between plants that hadthe same growth period overall during the experiment, theexperimental set-up included two short flooding treatmentsfollowed by a recovery period of different duration. Thelength of this re-growth period had no influence on survivalof plant species that originated from the river foreland,whereas it negatively affected some relatively intolerant andintermediately tolerant meadow species (Galium palustre andSilene pratensis as well as Arabidopsis suecica and Linariavulgaris; Fig. 2B). The increased mortality during recoverywas probably due to strong decay of the roots during the sub-mergence period, which was confirmed by a significant

Treatment

1. Control

2. Short floodingwith 2 weeksof recovery

3. Short floodingwith 5 weeksof recovery

4. Long floodingwith 2 weeksof recovery

1 2 3 4 5 6 7 8

Time (weeks)

Conditions

Drained Flooded

FI G. 1. Scheme of the experimental set-up testing the impact of flooding on survival and performance of selected plant species. Open bars represent growthunder control conditions, whereas grey bars indicate complete submergence.


positive correlation between root : shoot ratio of plant speciestreated with 3 weeks of flooding followed by 2 weeks of recov-ery, and their survival rates when allowed to grow for another3 weeks (5 weeks of re-growth) (R2 ¼ 0.84, P , 0.01).Furthermore, it was observed that in a number ofArabidopsis suecica plants, decay of the roots or the root–shoot junction during submergence caused detachment of theshoot, which floated up to the water surface and died.

Biomass

The results of two separate two-way nested ANOVAs on theeffects of treatment, habitat, species nested within habitat, andtheir interactions, and on treatment, tolerance level, speciesnested within tolerance level, and their interactions on plantbiomass parameters and plant height are presented inTable 2. During the course of the experiment, significantdifferences developed between species originating from thetwo habitats (Fig. 3). Species from river forelands subjectedto all treatments were characterized by a higher productionof total biomass, biomass of leaves, stems, fine roots and tap-roots than wet meadow species. The latter group, on the otherhand, had higher root : shoot ratios. When comparing groups ofspecies with different tolerance level to flooding, dry weightsof leaves and roots were higher in the tolerant than in the inter-mediate group, but root : shoot ratios showed an oppositeresponse.

The majority of tolerant species gained more leaf biomassthan was lost due to decay during flooding (e.g. Plantagomajor, Deschampsia cespitosa, Prunella vulgaris, Plantago

lanceolata and Rumex confertus; Fig. 3). This indicates thatbiomass production continued during flooding or commencedsoon thereafter. In case of Rumex crispus, the 5-weekre-growth period following short flooding resulted even in ashoot biomass comparable to that of the control plants,which grew for the same period (8 weeks) under non-floodedconditions. Lolium perenne, Achillea millefolium, Phleum pra-tense and Ranunculus acris increased their leaf biomass inboth short flooding treatments when compared with theinitial harvest, but their growth was negatively influencedduring 6 weeks of flooding. Among the intermediate species,Arabidopsis suecica, Galium boreale, Holcus lanatus andRumex acetosa increased or still maintained their leafbiomass during short flooding, whereas Cerastium fontanumand Linaria vulgaris were less sustainable to the biomass.Similar to the leaf biomass responses, tolerant species alsoincreased their stem biomass (Plantago major andRanunculus acris), and some of them actually reached thesize of control plants during the 5 weeks of recovery aftershort flooding (Deschampsia cespitosa, Plantago lanceolata,Prunella vulgaris and Rumex confertus). During and after sub-mergence, stem biomass did not change in two tolerant species(Phleum pratense and Rumex crispus), in none of the inter-mediate species (Arabidopsis suecica, Cerastium fontanum,Galium boreale, Holcus lanatus, Linaria vulgaris and Rumexacetosa) and in one relatively intolerant species Galium palus-tre (Fig. 3).

The taproot biomass of the river foreland species Plantagolanceolata, Plantago major, Rumex acetosa, Rumex confertusand Rumex crispus increased significantly in all treatments in

80

100A B

60

Sur

viva

l (%

)

20

40

0

Flooding duration (weeks)

0 1 2 3 4 5 6 2 5

Re-growth duration (weeks)

AmPl, Pm, Pp,Pv, Rco, Rcr

River foreland:

Lp

AsCfDc

Wet meadows:

GbGpHlLvRaSpVd

Am, Lp, Pl,Pm, Pp, Pv,Rco, RcrAsCfDc, Hl, Ra,RacrGbGpLvSpVd

FI G. 2. Relationship between (A) flooding duration and plant survival and (B) re-growth duration after 3 weeks of flooding and survival. The species abbrevi-ations are as listed in Table 1. Closed symbols, plants inhabiting river foreland; open symbols, plants from wet meadows; black continuous lines, tolerant species;grey lines, intermediate species; black dashed lines, relatively intolerant species. The initial number of plants was eight in the control treatment (except for Pv, Spand As, in which it was seven, and Hl and Gb, in which it was six), nine in short flooding treatments and eight in the long flooding treatment (except for As, in

which it was six, and Lv, in which it was nine).


comparison to the initial harvest, even after 6 weeks of sub-mergence (Fig. 3). Fine roots of most of the examinedspecies, however, were generally not able to grow well inthe flooding treatments. Exceptions were Plantago lanceolata,Rumex confertus, Rumex crispus and Plantago major, whichincreased fine root biomass including adventitious rootsduring or just after long submergence, although in the latterspecies 6 weeks of flooding caused stronger decay of theroot system, resulting overall in loss of root biomass. InAchillea millefolium and Deschampsia cespitosa, fine rootbiomass when subjected to short submergence was similar tothat of the initial harvest, but decreased after long flooding.Finally, roots of Arabidopsis suecica, Cerastium fontanumand Galium palustre showed decay rather than new growth.Other intermediate species maintained their initial rootbiomass during both flooding treatments, if they survived.

Shoot elongation

The maximum height of plants subjected to the floodingtreatments indicates the species’ ability of shoot elongation,which under favourable conditions can help to restorecontact of leaves with the atmosphere. In the current

experiment, none of the species was able to reach the watersurface of the 60-cm-deep basin, except for the tallest individ-uals of Galium palustre, Holcus lanatus and Linaria vulgarisimmediately at the start of flooding. However, these longerstem and leaf parts died within a couple of days and did notimprove survival of these plants.

The tolerant species Phleum pratense, Deschampsia cespi-tosa and the intermediate species Holcus lanatus elongatedtheir shoots to about 30–35 cm and thereby positioned theirleaves at a level in the water column with superior light con-ditions. Increased flooding duration resulted in a decline inheight (Fig. 4A), because of partial shoot decay. The onlyexception was Prunella vulgaris, a tolerant species, whichincreased in height to a similar extent during short and longsubmergence.

Almost all tolerant plant species (Achillea millefolium,Deschampsia cespitosa, Plantago lanceolata, Plantagomajor, Prunella vulgaris, Ranunculus acris, Rumex acetosa,Rumex confertus and Rumex crispus) elongated their shootduring flooding or in the re-growth period immediately there-after, when compared with the initial harvest. Surprisingly,also the intermediate species Arabidopsis suecica showedelongation of the shoot under water during the 3 weeks of

TABLE 2. Effects of (A) treatment, habitat, species nested within habitat and their interactions and (B) treatment, tolerance level,species nested within tolerance level and their interactions on total dry biomass, leaf dry biomass, stem dry biomass, taproot dry

biomass, fine root biomass, root: shoot ratio and plant height of the various species

(A)

F-values for:Treatment(d.f. ¼ 3)

Habitat(d.f. ¼ 1)

Treatment � habitat(d.f. ¼ 3)

Species(habitat)(d.f. ¼ 17)

Treatment � species(habitat)(d.f. ¼ 36)

Total dry biomass (g) 743*** 121*** 7.41*** 31.4*** 8.26***Leaf dry biomass (g) 50.6*** 503*** 15.2*** 76.2*** 5.70***Stem dry biomass (g) 584*** 412*** 58.0*** 45.3*** 6.78***Taproot dry biomass (g) 390*** 22.2*** 14.3*** 69.9*** 22.8***Fine roots dry biomass (g) 251*** 285*** 82.9*** 221*** 77.4***Root : shoot ratio 1081*** 464*** 41.3*** 49.0*** 7.26***Plant height (cm) 90.1*** 19.9*** 8.96*** 32.3*** 6.71***

(B)

F-values for:Treatment(d.f. ¼ 3)

Tolerance(d.f. ¼ 2)

Treatment � tolerance(d.f. ¼ 3)

Species(tolerance)(d.f. ¼ 16)

Treatment � species(tolerance)(d.f. ¼ 36)

Total dry biomass (g) 437*** 72.2*** 13.0*** 28.0*** 7.75***Leaf dry biomass (g) 51.0*** 371*** 17.1*** 47.8*** 5.37***Stem dry biomass (g) 402*** 189*** 43.4*** 62.4*** 8.04***Taproot dry biomass (g) 272*** 10.2*** 8.35*** 78.9*** 23.6***Fine roots dry biomass (g) 137*** 99.2*** 25.3*** 253*** 82.0***Root : shoot ratio 737*** 215*** 36.1*** 65.9*** 7.80***Plant height (cm) 79.0*** 19.7*** 13.5*** 33.8*** 6.25***

Effects were tested with two-way nested ANOVAs. Average values of the various traits are given in Figs 3 and 4. Level of significance: *** P , 0.001.

FI G. 3. Biomass of the various species after being subjected to different flooding treatments. The group of tolerance level is indicated in parenthesis after thespecies name: t, tolerant; m, intermediate; i, intolerant. The following treatments were applied: initial harvest, control growth for 8 weeks, short flooding for 3weeks followed by 2 weeks of recovery, short flooding followed by 5 weeks of recovery, and long flooding for 6 weeks followed by 2 weeks recovery, respect-ively. The columns present means of dry weight of the different plant compartments as indicated: above-ground biomass shown above the x-axis and below-ground biomass as negative values below the x-axis; d indicates that none of the replicates survived the treatment. Error bars indicate the standard error.Initial replicate numbers are shown in the legend to Fig. 2. The initial number of plants (species abbreviations as listed in Table 1) was eight in the controltreatment (except for Pv, Sp and As in which it was seven, and Hl and Gb in which it was six), nine in short flooding treatments and eight in the long flooding

treatment (except for As in which it was six, and Lv in which it was nine). Please note that panels have different scales.


–1

0

1

2

3

d

–3

–2

–1

0

1

2

3

4 A B C

D E F

G H I

J K L

M N O

–2

–1

0

1

2

3

d

–0·4

–0·2

0·0

0·2

0·4

0·6

7

–1·2

–0·8

–0·4

0·0

0·4

0·8

–1

0

1

2

3

d d

1·4

–5

–3

–1

1

3

5

d

–0·6

–0·2

0·2

0·6

1·0

–3

–2

–1

0

1

2

3

4

65 8

–2

–1

0

1

2

3

4

5

–3

–2

–1

0

1

2

3

4

2

–4

–2

0

2

4

6

4 2

Dry

wei

ght (

g)D

ry w

eigh

t (g)

Dry

wei

ght (

g)D

ry w

eigh

t (g)

Dry

wei

ght (

g)

–3

–2

–1

0

1

–4

–3

–2

–1

1

2

3

–3

–2

–1

0

1d

Achillea millefolium (t) Arabidopsis suecica (m) Cerastium fontanum (m)

Deschampsia cespitosa (t) Galium boreale (m) Galium palustre (i)

Holcus lanatus (m) Linaria vulgaris (m) Lolium perenne (t)

Plantago lanceolata (t) Plantago major (t) Phleum pratense (t)

Prunella vulgaris (t) Ranunculus acris (t) Rumex acetosa (m)

P Q R

Dry

wei

ght (

g)

3

2

3

2

3

–6

–3

0

–1

0

1

d d

–5

–4

–3

–2

–1

0

1

Rumex confertus (t) Rumex crispus (t) Silene pratensis (i)

Initial Control Shortflood + 2

Shortflood + 5

Long flood Initial Control Shortflood + 2

Shortflood + 5

Long flood

S

Dry

wei

ght (

g)

Plant compartments:

Verbascum densiflorum (i)

2

3

LeavesStemTaprootFine roots

–1

0

1

Initial Control Shortflood + 2

Shortflood + 5

Long flood

d d d


flooding treatment, although long flooding caused death of allplants. The remaining species, including the two relatively tol-erant grass species Lolium perenne and Phleum pratense, didnot increase their height (Fig. 4A), but observations duringthe submergence period indicated that also these plants (withexception of Silene pratensis and Verbascum densiflorum)had a more vertical positioning of their leaves, thus reachingcloser to the water surface.

As expected, the prolonged 5-week recovery after 3 weeksof submergence usually resulted in a height increase comparedwith the 2 weeks of re-growth period (Fig. 4B), but total heightwas still lower than that of the control group. Only three toler-ant species (Prunella vulgaris, Rumex confertus and Rumexcrispus) and one intermediate species (Cerastium fontanum)elongated very fast and reached or exceeded the height ofplants grown under control conditions for the same duration(8 weeks).

DISCUSSION

Irregular flooding can be an important stress factor for plantsthat inherently do not possess or are not able to developtraits enabling survival in submerged conditions. Especiallyspring and summer floods, which occur after strong rains(Blom and Voesenek, 1996), have a high impact on plant sur-vival (Vervuren et al., 2003; van Eck et al., 2004). Based onthe results presented here, it can be concluded that plantsfrom wet meadows are likely to be less tolerant to completesubmergence than plants from frequently flooded river fore-lands. All species from the latter habitat that were tested in

the present experiment were able to survive even 6 weeks ofcomplete submergence.

Wet meadow plants, which are continuously confrontedwith high soil moisture and temporary shallow soil flooding,probably possess traits that enable them to cope with oxygenlimitation in the root zone and with unfavourable biogeochem-ical reduction processes. However, deeper flooding, whenplants are completely submerged and oxygen, carbon dioxideand light availability are strongly limited, seems to require sol-utions at the whole plant level, including the shoot. Themajority of plant species from wet meadows could surviveonly 3 weeks of complete submergence. Two speciesgrowing at somewhat drier, sandier and higher elevated sitesin these meadows, Silene pratensis and Verbascum densi-florum, together with a species from wetter sites in themeadows, Galium palustre, did not even survive this shortflooding. The latter species is well-known as a true wetlandspecies, even though its roots are thin and do not contain aer-enchyma (Justin and Armstrong, 1987). It is likely that its rootstend to grow in the upper layers of the flooded soil, whichbecomes more oxygen-deficient during deeper floods. Theremaining two species from these meadows, Deschampsiacespitosa and Ranunculus acris, survived 6 weeks of completesubmergence and were classified as tolerant, although the latterspecies lost a large part of its biomass.

Enhanced shoot elongation is a well-known acclimation ofterrestrial plants to complete submergence that can restorethe contact of the leaves with light, carbon dioxide andoxygen. It can be a result of increased stem, petiole orlamina growth, and is accompanied by a vertical orientationof the leaves (hyponasty) (Laan and Blom, 1990; Banga

40

25

30

35

Am

Lp

Pm

Pl

As

Cf

Dc

Gb

Gp

10

15

20

Hei

ght (

cm)

Pp

Pv

Rco

Rcr

Gp

Hl

Lv

Ra

Racr

Sp

0

5


3 6 3 6


A B

FI G. 4. Height of the various species originating from (A) river foreland and (B) from wet meadows after being subjected to different flooding treatments. Thespecies abbreviations are as listed in Table 1. Symbols present means; error bars indicate standard error. Closed symbols, plants inhabiting river foreland; opensymbols, plants from wet meadows; black continuous lines, tolerant species; grey lines, intermediate species; black dashed lines, relatively intolerant species.Treatment replicates initially were five in the initial group (except for Hl in which it was four), eight in control (except of Pv, Sp and As, in which it wasseven, and Hl and Gb in which it was six), nine in both treatments with short flooding and eight in long flooding treatment (except for As and Lv in whichit was six and nine, respectively). Later, during the course of the experiment, the number of replicates depended on their survival during and after flooding.


et al., 1995; Blom and Voesenek, 1996). However, this type ofresponse requires considerable energy and carbohydrate input(Groeneveld and Voesenek, 2003) and it is beneficial onlywhen prolonged and/or relatively shallow floods occur. Forthis reason it was suggested that fast-elongating speciesshould be found rather in flood-prone environments withslow drainage and shallow temporal pools (Voesenek et al.,2004).

The present data indicate a significant difference in shootelongation between species from river foreland and thosefrom wet meadows (treatment � habitat effect on shootheight; Table 2A), which results particularly from the collapseof many meadow species after 6 weeks of flooding (Fig. 4).This is strengthening the results obtained by Voesenek et al.(2004), who subjected plants to the gaseous plant hormoneethylene to determine their potential for shoot elongation.Species from sites that were regularly flooded for longer dur-ation with shallow water depth showed stronger responsesthan those from habitats with different flooding characteristics(e.g. no prolonged flooding or only deep flooding). The presentstudy, however, considered the effect of complete submerg-ence on the plant as a whole, creating more complex stressconditions. Here too, shoot elongation appeared to be animportant feature of floodplain species, and even thoughthese plants were not able to restore contact with the atmos-phere, they placed their shoots in such a position that under-water photosynthesis was optimized. To a lesser extent, alsomany of the less-tolerant species showed such responses, par-ticularly by increasing the angle of the leaves towards a verti-cal position. This hyponasty may have shared responsepathways with shade avoidance, causing a constitutiveresponse under water even in non-wetland plants (Pieriket al., 2005), and, thus, potentially contributing to underwaterphotosynthesis. This trait may lead to an improved carbo-hydrate and oxygen status (Rijnders et al., 2000; Mommeret al., 2004) and is likely to be an important process allowingplants to survive. However, in previous studies it was shownthat flooding-tolerant species formed new leaves under waterthat are acclimated to submergence, whereas intolerantspecies did not possess such abilities (Mommer et al., 2006).Moreover, intolerant species did not have sufficient capacityfor internal gas transport, which makes it unlikely that under-water photosynthesis attributes substantially to survival ofthese species (Mommer and Visser, 2005). Evidence for thiswas particularly clear from Arabidopsis suecica, whichshowed complete disintegration of the root–shoot junctioneven though the leaves appeared green and viable.

Plant species from both types of wetland habitats showednot only large differences in survival upon complete flooding,but also displayed different patterns regarding the maintenanceof biomass and shoot elongation. Additional to survival rate,observed biomass changes can provide valuable informationabout the condition of the plants, their tolerance to total sub-mergence (Gibbs and Greenway, 2003; van Eck et al., 2004)and their ability to compete with neighbours afterde-submergence (Lenssen et al., 2004). Larger plants arelikely to have a competitive advantage over plants that justbarely survived the flooding period.

The present experiment showed that tolerant species weregenerally able to maintain or increase their leaf biomass

even during 6 weeks of flooding. In comparison to the tolerantgroup, intermediate species lost more of their leaf biomassduring the short flood, leading to a considerable reduction ofsurvival rate when flooding lasted longer. In some cases, leafbiomass actually increased during short flooding; however,these species all died during longer flooding, potentiallybecause biomass allocation to these leaves was costly and ren-dered too few benefits. The results confirm earlier data describ-ing similar patterns of leaf biomass reduction in relation totolerance to flooding with a limited number of species(flooding-intolerant Daucus carota, more-tolerant Rumexacetosa and tolerant Rumex crispus; van Eck et al., 2005). Inthis study, Daucus carota apparently was not able to accessits reserve of carbohydrates, probably because it lacked pro-duction of the specific enzymes required to mobilize carbo-hydrates during flooding, whereas Rumex acetosa and Rumexcrispus could profit from carbohydrate stores and thus slowdown the rate of decay of their biomass (van Eck et al.,2005). However, taproot biomass, being the main store ofcarbohydrates, did not specifically show decreases in the toler-ant species in the present experiment.

Since under flooded conditions, when most soil nutrients aremore readily available (Lamers et al., 2006), the presence of awell-developed fine root system is probably not the mostcrucial requirement for survival. Additionally, the mainten-ance of roots surrounded by hypoxic or anoxic soil requiresa continuous oxygen supply to these roots (Vartapetian andJackson, 1997). The tolerant species in our experiment(except for Achillea millefolium) from both habitats generallywere still able to enhance or maintain their root biomass duringflooding. This indicates that these plants were probably able tophotosynthesize under water or to take up oxygen via theirleaves (Mommer et al., 2004, 2005), and that they possessedan efficient within-plant system of gas transport (Laan et al.,1989; Blom et al., 1994). In contrast, most intermediatespecies lost more fine root biomass during 3 weeks of floodingthan they re-gained during the recovery periods. However, thisloss of root biomass had no apparent effect on plant perform-ance under water, except for when the decay of roots and theroot–shoot junction caused detachment of the shoot from theroots, as observed in Arabidopsis suecica plants. Much moreimportant is the role that roots probably play immediatelyafter the flooding period, when their condition and biomassgain will be decisive for plant survival. It is likely that substan-tial decay of roots can be the reason for plant death occurringup to 2 weeks after de-submergence, typically well after theperiod where post-anoxic injury due to radical oxygenspecies may result in reduction of survival rate (Nabbenet al., 1999). In the present experiment, soils were still quitewet in the first few days immediately after flooding, but laterbecame gradually drier. Under these conditions, the smallnumber of very short roots of the relatively intolerant speciesGalium palustre and Silene pratensis, and of the intermediatespecies Arabidopsis suecica and Linaria vulgaris, were notable to provide the plants with sufficient water.Consequently, plants with a low root : shoot ratio 2 weeksafter the short flooding generally showed lower survival ratesafter 5 weeks of re-growth.

Surprisingly, the patterns of total biomass development intolerant and intermediate species were not as predictable as


shown by van Eck et al. (2004). Tolerant plant species in thepresent experiment could either have high biomass loss fol-lowed by fast recovery in the re-growth period (e.g. Achilleamillefolium, Lolium perenne and Phleum pratense), or littlebiomass loss, followed by slower recovery. At the otherhand, although most intermediate species lost quite a largeamount of biomass during the short flooding, some of themmaintained or even increased their total biomass comparedwith the initial harvest, but still could not survive the longerflooding treatment. Apparently, total biomass developmentduring a flooding period does not necessarily predict the toler-ance of a plant species. Plants of certain species can survivelong periods of flooding with very small biomass (and havehigh recovery growth afterwards), and species that initiallymaintain a great biomass may suddenly collapse.

Based on the data presented here, it is concluded that ifformer river forelands, which were transformed into eitherdry or wet meadows, are reconnected to the main river bed(e.g. to function as temporary water retention basins duringhigh levels of river discharge), this would strongly influenceplant species composition and abundance. This effect will bemodified by submergence depth, duration, timing, frequency,water quality, turbidity and flow. Such conditions are likelyto favour species adapted not only to soil flooding but alsoto complete submergence. Flooding events in winter andearly spring or short summer submergence (,3 weeks) arenot likely to cause such dramatic changes, because they prob-ably will not eliminate the species that have an intermediateflooding tolerance. Much higher impact will have long floodsin the growing season, which may lead to the disappearanceof the majority of species present, and to disturbance of thebalance between any remaining species (e.g. in the case ofthe present research area, between the two co-dominantgrasses, the intermediately tolerant species Holcus lanatusand the tolerant species Deschampsia cespitosa, favouringthe development of the latter one).

ACKNOWLEDGEMENTS

We would like to thank Gerard Bogemann, Ronald Janssen,Gerard van der Weerden and his garden personnel for theirtechnical assistance, Heidrun Huber for her help with the stat-istics, and Zofia Stepniewska and Witold Stepniewski for theirsupport of this project.

LITERATURE CITED

Armstrong W. 1972. A re-examination of the functional significance of aer-enchyma. Physiologia Plantarum 27: 173–177.

Armstrong W. 1979. Aeration in higher plants. In: Woolhouse HW, ed.Advances in botanical research, Vol. 7. London: Academic Press,225–332.

Bailey-Serres J, Voesenek LACJ. 2008. Flooding stress: acclimations andgenetic diversity. Annual Review of Plant Biology 59: 313–339.

Banga M, Blom CWPM, Voesenek LACJ. 1995. Flood-induced leafelongation in Rumex species: effects of water depth and water move-ments. New Phytologist 131: 191–198.

Blom CWPM, Voesenek LACJ. 1996. Flooding: the survival strategies ofplants. Trends in Ecology and Evolution 11: 290–295.

Blom CWPM, Bogemann GM, Laan P, van der Sman AJM, van de SteegHM, Voesenek LACJ. 1990. Adaptations to flooding in plants riverareas. Aquatic Botany 38: 29–47.

Blom CWPM, Voesenek LACJ, Banga M, Engelaar WMHG, RijndersJHGM, van de Steeg HM, et al. 1994. Physiological ecology of riversidespecies: adaptive responses of plants to submergence. Annals of Botany74: 253–263.

van Bodegom PM, Grootjans AP, Sorrell BK, Bekker RM, Bakker C,Ozinga WA. 2006. Plant traits in response to raising groundwaterlevels in wetland restoration: evidence from three case studies. AppliedVegetation Science 9: 251–260.

Colmer TD. 2003. Long-distance transport of gases in plants: a perspective oninternal aeration and radial oxygen loss from roots. Plant, Cell &Environment 26: 17–36.

Crawford RMM, Brandle R. 1996. Oxygen deprivation stress in changingenvironment. Journal of Experimental Botany 47: 145–159.

van Eck WHJM, van de Steeg HM, Blom CWPM, de Kroon H. 2004. Istolerance to summer flooding correlated with distribution patterns inriver floodplains? A comparative study of 20 terrestrial grasslandspecies. Oikos 107: 393–405.

van Eck WHJM, Lenssen JPM, Rengelink RHJ, Blom CWPM, de KroonH. 2005. Water temperature instead of acclimation stage and oxygen con-centration determines responses to winter floods. Aquatic Botany 81:253–264.

Ellenberg H, Weber HE, Dull R, Wirth V, Werner W, Paulißen D. 1992.Zeigerwerte von Pflanzen in Mitteleuropa. Scripta Geobotanica 18:3–258.

Engelaar WMHG, Symens JC, Laanbroek HJ, Blom CWPM. 1995.Preservation of nitrifying capacity and nitrate availability in waterloggedsoils by radial oxygen loss from roots of wetland plants. Biology andFertility Soils 20: 243–248.

Fox GG, McCallan NR, Ratcliffe RG. 1995. Manipulating cytoplasmic pHunder anoxia: a critical test of the role of pH in the switch fromaerobic to anaerobic metabolism. Planta 195: 324–330.

Geigenberger P. 2003. Response of plant metabolism to too little oxygen.Current Opinion in Plant Biology 6: 247–256.

Gibbs J, Greenway H. 2003. Mechanisms of anoxia tolerance inplants. I. Growth, survival and anaerobic catabolism. Functional PlantBiology 30: 1–47.

Glinski J, Stepniewski W. 1985. Soil aeration and its role for plants. BocaRaton, FL: CRC Press.

Groeneveld HW, Voesenek LACJ. 2003. Submergence-induced petioleelongation in Rumex palustris is controlled by developmental stage andstorage compounds. Plant and Soil 253: 115–123.

Justin SHFW, Armstrong W. 1987. The anatomical characteristics of rootsand plant response to flooding. New Phytologist 106: 465–495.

Kundzewicz ZW, Ulbrich U, Brucher U, Graczyk D, Kruger A,Leckebusch GC, et al. 2005. Summer floods in central Europe –climate change track? Natural Hazards 36: 165–189.

Laan P, Blom CWPM. 1990. Growth and survival responses of Rumexspecies to flooded and submerged conditions: the importance of shootelongation, underwater photosynthesis and reserve carbohydrates.Journal of Experimental Botany 41: 775–783.

Laan P, Smolders A, Blom CWPM, Armstrong W. 1989. The relative rolesof internal aeration, radial oxygen losses, iron exclusion and nutrient bal-ances in flood-tolerance of Rumex species. Acta Botanica Neerlandica 38:131–145.

Lamers LPM, Loeb R, Antheunisse AM, Miletto M, Lucassen ECHET,Boxman AW, et al. 2006. Biogeochemical constraints on the ecologicalrehabilitation of wetland vegetation in river floodplains. Hydrobiologia565: 165–186.

Lenssen JPM, van de Steeg HM, de Kroon H. 2004. Does disturbancefavour weak competitors? Mechanisms of changing plant abundanceafter flooding. Journal of Vegetation Science 15: 305–314.

Lucassen ECHET, Smolders AJP, Roelofs JGM. 2000. Increased ground-water levels cause iron toxicity in Glyceria fluitans (L.). AquaticBotany 66: 321–327.

Lucassen ECHET, Bobbink R, Smolders AJP, van der Ven PJM, LamersLPM, Roelofs JGM. 2002. Interactive effects of low pH and highammonium levels responsible for the decline of Cirsium dissectum (L.)Hill. Plant Ecology 165: 45–52.

McDonald MP, Galwey NW, Colmer TD. 2002. Similarity and diversity inadventitious root anatomy as related to root aeration among a range ofwetland and dryland grass species. Plant, Cell & Environment 25:441–451.


Mommer L, Visser EJW. 2005. Underwater photosynthesis in flooded terres-trial plants: a matter of leaf plasticity. Annals of Botany 96: 581–589.

Mommer L, Pedersen O, Visser EJW. 2004. Acclimation of terrestrial plantto submergence facilitates gas exchange under water. Plant, Cell &Environment 27: 1281–1287.

Mommer L, Pons TL, Wolters-Arts M, Venema JH, Visser EJW. 2005.Submergence-induced morphological, anatomical and biochemicalresponses in terrestrial species: consequences for gas diffusion resistanceand phytosynthetic performance. Plant Physiology 139: 497–508.

Mommer L, Lenssen JPM, Huber H, Visser EJW, de Kroon H. 2006.Ecophysiological determinants of plant performance during submerg-ence: a comparative study among seven plant families. Journal ofEcology 94: 1117–1129.

Nabben RHM, Blom CWPM, Voesenek LACJ. 1999. Resistance to com-plete submergence in Rumex species with different life histories: the influ-ence of plant size and light. New Phytologist 144: 313–321.

Pierik R, Millenaar FF, Peeters AJM, Voesenek LACJ. 2005. New perspec-tives in flooding research: the use of shade avoidance and Arabidopsisthaliana. Annals of Botany 96: 533–540.

Rijnders JGHM, Armstrong W, Darwent MJ, Blom CWPM, VoesenekLACJ. 2000. The role of oxygen in submergence-induced petioleelongation in Rumex palustris: in situ measurements of oxygen in petiolesof intact plants using micro-electrodes. New Phytologist 147: 497–504.

Snowden RE, Wheeler BD. 1993. Iron toxicity to fen plant species. Journalof Ecology 81: 35–46.

Ulbrich W, Brucher T, Fink AH, Leckebusch GC, Kruger A, Pinto JG.2003. The central European floods of August 2002. Part 1. Rainfallperiods and flood development. Weather 58: 371–377.

Vartapetian BB, Jackson MB. 1997. Plant adaptations to anaerobic stress.Annals of Botany 79 (Suppl A): 3–20.

Vervuren PJA, Blom CWPM, de Kroon H. 2003. Extreme flooding eventson the Rhine and the survival and distribution of riparian plant species.Journal of Ecology 91: 135–146.

Visser EJW, Blom CWPM, Voesenek LACJ. 1996. Flooding-inducedadventitious rooting in Rumex: morphology and development in an eco-logical perspective. Acta Botanica Neerlandica 45: 17–28.

Voesenek LACJ, Rijnders JHGM, Peeters AJM, van de Steeg, de KroonH. 2004. Plant hormones regulate fast shoot elongation under water:from genes to communities. Ecology 85: 16–27.

van der Welle MEW, Roelofs JGM, op den Camp HJM, Lamers LPM.2007. Predicting metal uptake by wetland plants under aerobic andanaerobic conditions. Environmental Toxicology and Chemistry 26:686–694.

Wolsink M. 2006. River basin approach and integrated water management:governance pitfalls for the Dutch Space-Water-Adjustment ManagementPrinciple. Geoforum 37: 473–487.


Differences in flooding tolerance between species from two wetland habitats with contrasting hydrology: implications for vegetation development in future floodwater retention areas

Documents