Jacquemart, Anne-Laure ; Champluvier, Dominique ; De Sloover, Jacques. Document type : Article de périodique (Journal article)

Description
A test of mowing and soil-removal restoration techniques in wet heaths of the High Ardenne, Belgium Jacquemart, Anne-Laure ; Champluvier, Dominique ; De Sloover, Jacques Abstract Vegetation restoration

Please download to get full document.

View again

of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Paintings & Photography

Publish on:

Views: 49 | Pages: 11

Extension: PDF | Download: 0

Share
Transcript
A test of mowing and soil-removal restoration techniques in wet heaths of the High Ardenne, Belgium Jacquemart, Anne-Laure ; Champluvier, Dominique ; De Sloover, Jacques Abstract Vegetation restoration was attempted in four plots of a degraded wet heath overgrown by Molinia caerulea in High Ardenne (Belgium). Two treatments, mowing for two years and removal of the top 3-28 cm of peat, were applied in spring The vegetation was recorded in 1989 and 1998 using the Braun- Blanquet method. After nine years, the mowing treatment had relatively little effect on vegetation composition. We observed a 0-25 % decrease of Molinia cover and an increase of some heath species like Calluna vulgaris (12-47 %) and Erica tetralix (0-12 %). However, top peat removal had a pronounced and beneficial effect. A decrease of a mean of 66 % in cover of Molinia caerulea was observed as well as an increase of the cover of Erica tetralix (0-12 %). Peat removal also promoted the re-establishment of some typical wet heath species like Juncus squarrosus and Carex panicea, as well as of some rare pioneer species like Drosera rotundifolia and Lycopodiella inundata. This treatment appears... Document type : Article de périodique (Journal article) Référence bibliographique Jacquemart, Anne-Laure ; Champluvier, Dominique ; De Sloover, Jacques. A test of mowing and soil-removal restoration techniques in wet heaths of the High Ardenne, Belgium. In: Wetlands : the journal of the Society of Wetland Scientists, Vol. 23, no. 2, p (2003) DOI : / (2003)023[0376:ATOMAS]2.0.CO;2 Available at: [Downloaded 2016/10/10 at 15:23:31 ] WETLANDS, Vol. 21. No. 2. June 2003, pp E5 O Thc Society ol wetland Scienti'r!\ A TEST OT'MOWING AND SOIL-REMOVAL RESTORATION TECHNIQUES IN WET HEATHS OF THE HIGH ARDENNE. BELGIUM Anne-Laure Jaccluemnrtr. Dominique Champluvier'], and Jacques De Sloover tunité cl'ecologie et de Biogéogrttphie Cetûre de recherche sur la biodiversité U n i' e r.s it é cat holique tie Louv aîn 5 Place Croix-du-Sud Belgittm Louvain-h Neuve E nail: ecol.ucl.dc.be ) National Botantcttl Garden of Belgium Domairte de Bouchout Belgium Meise Abstrdct: V etetl.tion restoration was attemptcd in four plors of â dcgrâded wct hcath overgrowi by Molitlia caerulea in High Ardenne (Belgium). Two treatmerls. mowing fbr two years and removal of the top 3 28 cm of peat. were applied in spring The vegetation was lecorded in 1989 aud 1998 usiug lhe BrauD Blanquet method. Aftcr nine years. the mowing t.e,ltmerlt had reiatively little eftcct on vegctation composition. We observed a decreasc of Mctlinia cover and an incrcasc of somc heath species like C.7//r/ra? t'ulgaris (I24'/ L/o) and È,rit:tt!?tralix (O l2 7.). However. top peat reûiolnl had a pronounced and bcneficinl etïect. A decrease ol a inean of 66 9i, in coref ol Malinid cd?ntlea wàs ohserved as wi:ll âs an increase of the cover of Éi-r.i7 tetrdlir (O-12 7.). Peat removel âlso promoted the rc establishmedl ()1 soirc typiçal \\'et heath species \1ke Junt:us squarrotûs n l (:(1rd pani( d.r- as well âs of some rare pionccr spccics likc I)rcsc1a tottudiloliù and L copodiella inunllata.lhrs treatrnent appears to be iavorable for the regederation of degrlded \\'et heaths in lligh Ardennc. (er l!'rrrrly; degraded wet heaths, top peat rcmovîl- Molinia tderulea. wet heâth restoralion IN ITODUCTION Vegetation 'restorntio l' by its restdcted cletinition implies a letllin to a tirrmcl or-oliginal state (Pfadenhauer ancl Klijtzli Higgs 1997). Thc degree of sidlilarity between l lb rner (or',rriginal) and a restorcd state is not cxplicit. but it is inrplicit that a restored object shoultl havc substanlial sinlilîrity rvith a former condilion (Briilisaùer and Kkjtzli 1998). This paper investigated the use of peat removîl and. to a lcsser extent. mowing as teclrniques tbr restoring of wet heaths in the platearl des Tailles. in Upper Ardenne. Belgiunr. Bogs unrl \\,ct heaths occur predominantly in the cool temper-ate zorres of Eurasia ancl Ancûca and uere once lnuch rnore $idespread in Europe than they arc k)day. Ilistclrically, losses in Europe rvere due to the conversion of heaths for-ii-lr-iculture and forcslry (Gôttlich and Kiiule Aert\ et.rl. I995- Wheeler and Shai',, Brooks and Stoneman 1997). Until the beginning of past ccntury. largc areas in fjpper Aldcnne were covered by \\,et lreâths. This semi natural vegetation occupied thin peaty soils (less than 50 cn) and u,as maintâined bv traditional agri culture (Cardoen and Mrftens I)unront arrrl Charnpluvier 1990), such as extensiye shcep grazing. mowing, and rye (.\'.,.l]1t, t ert,ule 1,.) cultivi.ltior (Heyois 1981). Rye fire cultivation was prlcticed in Ardcn nc duûng the eighteenth and ninetcenth centuries. This consisted of superficial drrining and llurnin-g.lf thc upper peat Iayers lirllowed by cultivatior tl1 r)'e in the resultant âsh-charcoal pelrt mixlure. Now, rnady wet heaths occur on the rnargins of mires. as traclilional peat cuttirg has atlècted nearly all thc bogs in the area r Ho\,,i. l ()8 1 ). Using tlre European Braun Blancluet classilication. three peaty heath corrmunities are recognized en the plateau des Tailles, in Upper Ardenne (De Sloover et al. 1980): (l) the O-tlcot co Eric luttt /Earolici.s association belonging to the O-rr..)( (?) Splugneter. (2) the slr b- rssociatien ti the Callurto Nartlclett with Erictt te tl-.r/r.r of the Vattittirt t,itis idacttt' CullLurctunr: and (3) also in the Culltr o-nurdelea association. the Jù11(:o s.ludrt osi-nurdellrl, û scmi bushy heath. Once agricultural management ceases. these wet heaths rapidly conve.t into mono-dominant Molinl.r.?r?rr1..r grassland (Chanbers et al Taylor et 376 Jacquemart el zrl., VEGETATION RESTORATION IN WET HEATHS 3'77 al. 2001). This change is generally ascribed ro ir]- creased nutrieûî availability (particularly nitrogen) in these nutrient poor ecosystems (Dumont and Chanpluvier 1990, Berendse et al. 1994, Bakker and Berenrlse Chambers er al. 1999). Molinia caerulea (L.) Moench is an erect, compactly tufted pcrennial grass, 15 to 250 cm high when flowering, forming either tussocks or e[tensive swards. This species is a common native species, widespread in Europe, usually abundant, and frequently dominating large areas, ofien to the exclusion of other flowering plants (Taylor et al. 2001). Species diversity decreases because rnany higher plant species, bryophytes, ând lichens decrease in both species number and in cover (Chambers et al. 1999, Taylor et al. 2001). Many insect species, rare reptiles. and bird species (e.9., Tetruo rernr L.) also have disappeared from such sites in Belgium (Frankard et al. 1998). These Belgian cor-nmr.rnities evçntually undergo succession to forest, after invasion by Szr/in aurita L., S. )1 mttl.tineruis DiiIl., S. caprea L., Sorbtrs tlucuparia L., BetuLa pulre.scens Ehrh., and QLlercus roàar L. (Dumont and Champluvier 1990). As a consequence of thcse impacts, species-rich wet heaths are threatened ecosystems. Their restoration oftcn requires a reduction of nutr-ient availability (M;urs 1985, Man's and Gough 1989, pywell et al. 1994, Aerts ct al. 1995, De Graaf et al. 199g. Bakker and Berendsc 1999). Restoration success depends strongly on species establishr.nent (e.g., diaspore bank and dizrspore dispersal) and species survival in relation to soil and water conditions (Putwain and Gillham 1990, Mitchell et al , 2000). Some resro ration experirnents harve been carried out on degraded wet hcaths and tricd to decrease nutrient availiibility (Diemont et al l)iemont and Linthorst Hor.nan Cardoen arnd Martens Marrs 1993, Pywell et al Aefts et al. 1995, Moen Nolan et al. It)95, Dnnne and Doyle 1998, Mitchell et al. 1999). One way to remove excess nutrients is to rernove top soil (Werger et al. 1985, Marrs Sansen ancl Koedaur 1996). The depth of the layer ro bc removed de pcnds on tlte nutrient concentrittion at diiïerent soil depths. Scraping to a depth of c. 30 cm may be sut'- ïicient in most cases (Dienont Mans Mitchell et ai.2000). This rreatment is drastic. and it is often not known how this removal affècts macronutrient availability or what proportion of the soil seed bank would remain (Marrs 1993, Dupieux I 998). Another metl]od for reducing nutrient availability is mowing (Marrs 1993), which has been successlully tested in Norway (Moen 1995). Increased mowing liequency and ren.roving the cuttings would obviously remove more nutrients fiom the soil than one cut per year would, but this technique may weaken some of the characteristic species of fen meadows, particular-ly species that flower late in the year, such as Molinin caerulea. Grazing could also be used successfully (No lan et al. 1995, Dunne and Doyle 1998, Dupieux 19913) if appropriate shepherding is available, which is nor the case on the plateau des'l'ailles. As the techniques 1-or wet heath rcstorarion âre poorly documented (Dupicux 1998). the aim of this study was to investigzrte thc cffects of two treâtments: (l) surface peat removal (up to the rrideritl layers) and (2) two years of mowing, oû the floristic composition of degraded wet heaths. The vegetation was surveyed nine years after experimental treatments were applicd. The objective of these treafments was to red\tce ML'Iinia coyer and promote the re-establishrnent of typical wet heath species. Evcn with only two yeafs of mow ing, we cxpected to reduçe Moli/ri.? tussocks and pril mote other species. Site I-oc:lti()n METHODS The study was conducted ncar Vielsalm at the plateau des Tailles in the Uppcr Ardcnne, Relgium (50 14' N. 5 47' E, m alt ) Thc sitc was located ill the nature reserv(] 'Réservc Naturelle Domaniale de la Grande Fange de Bihain' of approx. 200 ha (Figure l). The soil prolile consisrs of a 5- to 30 cm layer of line compact black peat containing rhizomes and roots, overlaying a cm minerll horizon of light brown compact sand and silt.'l'he mei:ln annual rainfirll is about 1260 mm. and the ûrean annual ternperaturc is 6-9 C. During parst cerrturtes, the sile was used for peat digging and sumûler grazing, as well a5 ryc cultivation- At the beginning of the past ccntury. a significant part of this area (dry to wet) was drainell and planted with Picen abias (L.) Karsr. Today, the vegetiltion of a great part of the Reserve is do[lirated 6y Moli.niu degruded wet hcath llntl Piceo plrntations. F rperiment. In May 1989, tbur plots (15 X 10 m) were chosen. They represented highly degraded wet heaths, dominated by MolirLicr. Plots were located on the margins of the reserve. and so they were easy to reach with machines (Figure I ). In each plot, three quadrats (5 x l0 m) were delim ited with wood stakes. and the fbllowing treatments were appliecl: (l) mowing (Ml to M4), (2) removal of surfacc peat (Pl to P4), and (3) unmanaged (Ul to U4). Peat renoval was caried out on May 29 and 30, 1989 with an excavator. Only I 2 cm of peat was left above the mineral soil. Mowing was done at the end 378 WETLANDS, Volume 23, No. 2, Belgium A Figure 1. Location of the plateau des Tailles, Belgium (A) and of the experimental quadrats in lâ Gratde Fange Reserve (B) wet heaths (U: unmanaged, P: lop peât removed, M: mowed). of the spring in 1989 and 1990 (see Dumont and Champluvier 1990 for details). Vegelation Survey and A nalyses A total of 16 relevés: 4 (I1-I4, one per plot) in 1989 before restoration treatments and 12 (one per quadrat) in 1998, following the Braun-Blanquet method, were available for analysis. Nomenclature follows Lambinon et al. (1993) for Spermatophytes, Daniels and Eddy (1990) for Sphagna, vanden Berghen (1979) for liverworts, and Smith (1978) for other Bryophytes. The species data were classified by Ward's method (agglomerative hierarchical fusion) using SIMIL (Steinhaus similadty index) and CLUSTER using the 'R Package', following van der Maarel ordinal transformation of the cover-abundance scores (Legendre and Vaudor 1991). Principal coordinates analysis (PCA) was also used (SIMIL and PCOORD, Legendre and Vaudor 1991). The program IndVal was used to determine the indicator species of the groups (Dufrêne and Legendre 1997). This IndVal index value is maximurn (100 Vo) when all individuals of a species are found in a single group of sites ( specificity ) and when the species occurs in a1l sites of that group ( fidelity ). The statistical significance of the species in dicator values was evaluated using a randomization procedure (1000 bootstraps, p : 0.05). Spearman correlation tests were performed between species number and peat depth (SAS Institute 1993). Jacquema't et el., VEGE'IATION RESTORATION IN WET HEATHS yt tt tt tt 1.2 I 't.4 I CI o N o (9 M4 -M3?[ '. r L o,o, Figure 2. Pcoord classification of quridrat samples from La Grande Fange wet heaths. Symbols of samples refer to treîtments (I: initial stâge in 1989, U: unmanaged, P: top peat rcmoved, M: uowed). Noda are iùdicated. Vegetation Classillcation RESULTS Using vegetation classification generated by Ward's rnethod (Figure 2, Legendre and Vaudor 1991) and PCA resr.rlts (Figure 3), a three-nodum level was chosen after randomization to maximize Indval values. PCA axes 1, 2, and 3 accounted for 30.8, \'1.1, and 14.l o/a of the floristic variation, respectively. The main diffèrence was between plots 1-2 and plots 3 4. The vegetation of plots I and 2 was distinct (Table 1) and occupied different parts of the PCA ordination (Figure 3). Within these two groups, the mewn quadrats were grouped with the unmanaged ones. The second main difference separated the quadrats where peat was removed from thc unmanaged and mown quadrats (Figurcs 2 3). According to Indval vâlues (expressed in percents),. the following noda were recognized (they represcnt a sequence fi'om unmanarged to quadrats with peat removal). (1) Northecium ossifragum rodum (ll, Ul, MI), with a high constancy o[ Narthecium ossifrugum att,d Ericaceae species, and a high number of mosses and liverworts (.e.9., Calliergon stramineum, Calypogeia fissa, Campyloltus fexuosus, Cephalozîa lammersiana, Clodctnia portentosa, Hypnum cupress(itrme, H. ju tl.andicum, Le uco b ryutn g la ucum, Lo p ho zio b ic re nara, and Polytrichum stricrum). Except for the presence of Narthecium, this group was quite similar to plol2 (I2, 380 WETLANDS, Volume 23, No. 2, 2003 û!;=' ;t e*èj Ëir ô = il i 3 à,: É ù _ dd d ô-,ô.\-!n!.- - i-ô,- d- 6 è ê -ri aï at g â!:-q: ql F{.È:i é i= F É ç:.5: gi.e :-l!s s 'Ë - =r6 - Èilf-- læcc- lr,cr,c N -*i = r s 3-: t q:. s.- Ê.-; o - , -:-. ;r ù :- I F -i.-*.c F z --ô -* ' ddo-d * *- I ^c, o c! q Ll * ÈS qù:; c- n:4= 3 rùs û! i i = o I ] S- '.3;-: ai--' : 9.r. i:. Y ù i 9 o : :3 =-='-! -. =^=aà Ç ^ = :: E -. 1 i. * r î: j a À : ; i- -:=: i j q ;n : 7 -.-:È:-r:-è : : I zl i: -= - r ! : t :,.j : =1 i : LÈâ.-i 47 {:ii!:i.ti ;E::1:l'ii:ir::! .,.3:S:-:til: f 5 s s - i ; :.S ::}-^--.e 3È: ,*a::!3,?\::: : è : : : :! 3!;,;1--,;i i: l::.-4 -: ;!'ÈÀ= s i 7 S:::i!: Ê :.-=.: i : i É : S I Sr ii : : 3 :!: t^i ; : { :-ri i È q: t -q= i c:: =' :î, È Ë : i è AÊi I L i:-è.è.: à -:.È.if - i - 9! t o è - { i i i: i \ : \ : ' cz *l; U Q -- âç \J ! J /t ::1 ÊE:; Ài:E ièè.âét:ts! ii I?'=2--l I..ri d= âi2;2: : È È Jacquemart et al., VEGETA'IION RESTORATION IN WET HEATHS 381 c!ni ' êi -icr *dê ;o ( - * '1 * '..1. E F z. -ri - O 9 'Ë 1 qÈ-td âri Ë È Ë ,.i e E Ë,,= É=8 = ë Èç ië= zz,: =;;: -; â. i... i : E i= t!ë * reésai ç t.= r ?, - ijé:-.!èi:è,={,:= 4, z E ËS:ÂËi ;i pè;è!=:ë;3s:: s i:,:; ir-;'èèiiè S! :i :Èir;ÈÈ ÉiÈËi:r?i::!. l;5è:!i:èiè É ll=siè tg:tè! é :: y,iià-:i i- i:3ie = sstsè i:s:; E-ÈÈ;ÈÈ:iÈtJ: :: Ë$-Èi$i;:*_È::È Ë!:5{: s::=:èë:i.r9 -' Ë i,sèéàâ I sàè-ri; g ss:ë Ë ÊÈ$;ièÀFÀè;.i -i o 7 i 382 WETLANDS, Volume 23, No. 2, 2003 P2 U1 P3 t\,41 M2 Ir,43 4 Are 2 --- Axe 1 Figure 3. 2-D ordination of quadrat samples along principal components axes I and 2 (30.8 and a respectively of the total floristic valiance). Symbols of samples rc1êr to treatments (l: initial stage in 1989, U: unmanaged, P: top perl rcmovcd, M: mowed). U2, M2) which presented the fewer number of species (5 to 12) of a1l quadrats. The indicator species using Indval were Deschampsia f.exuosa (77.2 o/o), Vaccinium yitis idaea (.76.9 Vo), Vaccinium myrtillus (68.9 7o), Scirpus cespitosus (53.8 Eo), Narthecium ossifrtt gum (5O.O Vo), and Vaccinium uliginosum (47.4 Eo). (2) A second gloup was also dominated by Molinia caerulea (.13, U3, 14, U4, M3, M4) but presented morc species, with Viola palu.stris (81.2 Va), Dryopteris cctrthusiana (61.8 Va), Angelicct sylyestris (51.3 o/o), and Polytrichum commune (5O.9 o/o) as indicator species. The mern height ol herbs rnd grr\\es wc\ more impoftant than in the plots 1 and 2 (Table l). (3) The four peat-r'emoval quadrats (Pl-P4) were well-separated from the others:./rrr?ca,r squdrrosus (91.7 Eo),.1. bulbosus (61..4 Ec), Carex clemissa (91.3 o/o), C. panîcea (87.8 Eo). Sphagnum cepillifol.ium 1)ar. capillifolium (65.9 %), ard S. auricukûum (58.7 9o) were characteristic (Indval values ) 50 Vo'1. Molinia caeruleo coyet was reduced (abundance of 4 or 5 in 1989 in Il-4, 1 or 2 in 1998 in P1-4), and some pioneer species were found only here (e.g., Drosera rotund{ôlia, Lycopodiellct inundata, Sphagnutn nngellanicum, and S. subsecundum (Table 1)), Variat ion in Vegetation Composirion Species richness of the vegetation samples ranged from 5 to 27 species per quadrat. The total number of species reached 69 in 1998 compared to 4zl in There were 42 specics in the unmanaged plots in 1998, 42 species in the peat-removal quadrats, and 40 species in the mowed quadrats. The number of species was not correlated with treatment or year but was cortelated with peat depth (before tr-eatment), with more spe cies in plots on deeper peat (r : 0.728, a : l8). Between 1989 and 1998, the vegetation evolved without any direct anthropogenic intervention. -/rr?cils bulbosus, Polytrichum strictum, ànd Sphagnum papil lasam disappeared. Scirpus cespitosus, Vctccinium vi tis-idaea, V. oxycoccos, Deschampsia flexuosa, and S. capilli-folium var. capillifolium decreased, whereas Calluna vulgaris and Cirsium pelustre increased. It is noticeable that low densities of several tree species JucqueDrart.r./1., VECETATION RESTORATION IN WET HEATHS 383 appeared or developed in all plots (seedlings of Picezi abies, Pi.ttu,s,ry1ve.sll1.r, Sttl.ix uurittr, S. \ nultinervis, ancl Sorlras aucupario'). For exampie, a single Sdàr X ntultirten,is bush developed otr several square meters in [J2 quatlrat. Effects of Mowing Difïèrences between the mowed treatment and the unmanaged quadrats were quite minor (Figures 2-3). The effect of mowing was not drastic: however, a slight decrease of Molinia caerulea was observed i[ quadrats 3 and 4 (abundance of 5 in 1989 and 4 in 1998) but not in quadrats 1 and 2. Vctccinittm n)-rtill.us and Epilobium.ùlgustifoliuÛl disappeared (Table 1). On the other band, Nanheciunt oss(tttgum. Call.unu vulgaris. Eriophorum pol)stqchion, Care.x echinata. Vttccittiurrt t'ilis-iduea, Ericd tetralix, and Sphagntun polustre increased. Two new species were detccted: Dac lyl o rltizu,sltln gnit-ola a:od J unclls.rql!arro.t u s. Effects of Peat Removal Nine years after treatments, bare patches were found in peat-removal plots, although the total cover of vegetation reached between 55 and 99 Va. The bare patch es (all less than 5 m,) were located in the wettcst places, which were undef water for several months per yein. Nevertheless, there was a decrease in Molin.io c.rcrule.l coyer, as well as Cirsium palustre and. LeLr, cohrytun gl.aucunl, and a loss of Vaccinium ûtis-idaea. V. nyttillus, Dry-opteris cartll.usian.i, and Epilobium 0ngLrr tifoliun (Table 1 ). However, pioneer species sltch as Sphagnun auriculatun, S. palustre, S. recurvum, and Agro.stis canina increased. Carex demisstl and JLotcus ejflrsrs increased or appeared in thc stripped areas. Carex ponicea, C. pilul(èra, Juncus sqtlarrosus. J. bulbosus, Lytrtpodiella inundata, Drosera rotttndi
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x