School Work

Smoke, Mulch, and Seed Broadcasting Effects on Woodland Restoration in Western Australia

Smoke, Mulch, and Seed Broadcasting Effects on Woodland Restoration in Western Australia
of 10
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.
Related Documents
   JUNE   2002  Restoration Ecology Vol. 10 No. 2, pp. 185–194  185  © 2002 Society for Ecological Restoration  Smoke, Mulch, and Seed Broadcasting Effects on Woodland Restoration in  Western Australia   Deanna P. Rokich   1,4  Kingsley W. Dixon   1,3  K. Sivasithamparam   2  Kathy A. Meney    1  Abstract  Smoke, canopy-derived mulch, and broadcast seedswere used to maximize the establishment of Banksia  woodland species in sand quarries in Western Austra-lia. Smoke, particularly aerosol smoke, had a positiveeffect on total seedling recruitment. Pre-mined (wood-land) sites showed a 42-fold increase in total germi-nants and a 3-fold increase in the number of specieswith aerosol smoke application. Post-mined (restored)sites showed only a 3.6-fold increase in total germi-nants and a 1.4-fold increase in the number of species.Two water-based smoke chemicals, DC10 (pH 4.5) andSC63 (pH 2.5), increased seedling recruitment at bothsites but were not as effective in stimulating recruit-ment as aerosol smoke. Neither of the chemicals wereeffective in significantly increasing species richness. Application of aerosol smoke directly to seeds as a pre-treatment before broadcasting had no effect on seed-ling recruitment. Broadcasting of seeds onto restora-tion sites significantly increased seedling abundanceand richness. Application of a single layer of mulchfrom the canopy vegetation after seed broadcastinggave optimum seedling recruitment. Two layers ofmulch significantly reduced recruitment, as did ap-plying mulch before seed broadcasting. For broad-scale restoration, the application of smoke on newlyrestored sites would be more effectively achieved us-ing smoke water sprayed over the soil surface. Speciesthat do not recruit from replaced topsoil could be ef-fectively recovered from broadcast seed rather thanfrom mulch.  Key words: aerosol smoke, Banksia  woodland, germi-nation, mulch, sand extraction, seedling recruitment,smoke water, species richness, topsoil, restoration.  Introduction  T  he Banksia  woodland community is widespreadand common on the Swan Coastal Plain in thesouthwest of Western Australia. Rocla Quarry ProductsPty. Ltd. operates a sand extraction quarry on Banksia  woodlands over deep siliceous sands 30 km northeastof Perth, Western Australia. The extraction process in-volves removal of the Banksia  woodland vegetation and soil profile to great depths (18–22 m). The company pro-poses to restore post-mined sites with an ecosystemclosely resembling the pre-mined plant community. Un- fortunately, the recruitment biology of Banksia  woodlandspecies is not well understood, providing limited oppor-tunities for devising appropriate prescriptions for resto-ration and management of this plant community.As a routine part of the restoration process the com-pany removes the woodland topsoil, which containssoil-stored seeds, and returns the topsoil into areas to be restored. The value of fresh topsoil for increasingplant diversity in mine-site restoration is well docu-mented (Tacey & Glossop 1980; Koch et al. 1996). Forthose species that do not store their seeds in the soilseedbank, replacement of species is achieved by use of  broadcast seeds or through planting of green stock intorestored sites. A mulch comprising canopy material of the native vegetation is occasionally applied to areas being restored after topsoil replacement. Followingthese practices, the diversity of recovered Banksia  plantcommunities has been limited (S. Elliot and S. Fletcher,personal communication, 1995).The recent development of aerosol smoke as a germi-nation tool (Dixon et al. 1995; Roche et al. 1997, 1998)has provided opportunities for increasing recruitmentand species biodiversity in mine-site restoration prac-tices. The application of aerosol smoke, however, is notconsidered effective or practical for use in broad-scalerestoration activities. After their initial smoke studies,Roche et al. (1997) tested the application of smoke water   1  Kings Park and Botanic Garden, West Perth, WA 6005, Aus-tralia 2  Soil Science and Plant Nutrition, Faculty of Natural and Agri-culture Science, The University of Western Australia, Ned-lands, WA 6907, Australia   3  Plant Biology, Faculty of Natural and Agricultural Science, The University of Western Australia, Nedlands, 6907, Australia   4  Address correspondence to D. P. Rokich, Kings Park and Bo-tanic Garden, West Perth, WA 6005, Australia. Tel.: 08 94803640; Fax: 08 94803641; E-mail   Smoke, Mulch, and Seeding Effects on Woodland Restoration  186  Restoration Ecology  JUNE   2002  to the soil seedbank, a more practical method of apply-ing this germination trigger over large areas. There wasa significant germination response, with total germi-nants and species numbers increasing by 56 and 36%,respectively. Additionally, Roche et al. (1997) reportedthat treatment of seed mixes with aerosol smoke before broadcasting resulted in highly significant increases intotal in situ recruitment and species numbers whencompared with unsmoked seeds. Interestingly, evenspecies normally unresponsive to smoke ex situ wereresponsive to smoke in situ. It was suggested thatsmoke may act as a deterrent against in situ seed preda-tion (K. Dixon, personal communication, 1997). In addi-tion, these authors found that smoke residues carried by the smoked broadcast seed may stimulate germina-tion of seeds already resident in the soil seedbank.Maximizing rehabilitation of the full suite of speciespresent before mining requires an understanding of theecology of the species involved, particularly their seedstorage syndromes. Although the use of the topsoil seed- bank can contribute the most seedling recruitment inpost-mined sites, many species store their seeds in per-sistent fruits held in the vegetation canopy (serotiny). Aunique feature of the kwongan vegetation (a fire-adaptedshrubland) in Western Australia is that a substantial pro-portion of the seeds are stored in these persistent fruits(Bellairs & Bell 1993). Bellairs and Bell (1993) reportedthat a mulch of the canopy material was the most impor-tant source of seeds in post-mined sites, containing85% of the total germinable seeds, whereas topsoil and broadcast seed contributed 13 and 2% of the germinableseed. On the contrary, at Alcoa of Australia’s mine sitesin forest areas in the southwest of Western Australia, broadcasting of native seeds provides one of the bestmethods of achieving high plant densities and cover (Koch & Ward 1994). It has been proposed that mulchmay also ameliorate adverse conditions affecting seed-ling recruitment by decreasing wind and water erosion(Bellairs 1992). It would appear that the value of thetopsoil, broadcast, and mulch-based seed varies withplant community, topsoil handling and storage proce-dures, mining type, practicality, and cost effectiveness.This study represents part of a broader investigationon plant ecosystem restoration of Banksia  woodlands.Specifically, we investigate (1) the effect of aerosol smokeand smoke water on the seedling recruitment from thetopsoil seedbank, (2) the effect of aerosol smoke on broad-cast seeds as a deterrent against seed predation, and (3)the optimum concentration of mulching and order of mulching and broadcast seeding on seedling recruit-ment.  Methods  Study Site  The study sites were located within an area previouslymined by Rocla and an adjacent Banksia  woodland as areference area (Fig. 1). The Banksia  woodland comprises afloristically rich and taxonomically diverse plant com-munity similar to the adjacent kwongan plant communi-ties in the northwest of the Swan Coastal Plain (Dodd &Griffin 1989). The tree canopy of the Banksia  woodland iscomprised of Banksia attenuata (  slender banksia)   and  B.menziesii  (firewood banksia),   with Eucalyptus todtiana(  prickly bark) and  Nuytsia floribunda  (Christmas tree)occurring less frequently (Beard 1989; Dodd & Griffin1989). The understory is represented by members of the dominant woody families Myrtaceae, Epacridaceae,Proteaceae, and Papilionaceae, with dominant herba- Figure 1. Banksia  woodland before quarry-ing activities.   Smoke, Mulch, and Seeding Effects on Woodland Restoration  JUNE   2002  Restoration Ecology  187  ceous elements represented by members of the Antheri-caceae, Stylidiaceae, Haemodoraceae, Cyperaceae, andDasypogonaceae families (Trudgen 1997). The Banksia  woodland habitat experiences a typically Mediterra-nean-type climate of cool mild winters and hot drysummers and is characterized by deep, well-drained,nutrient-poor, sandy soils.  Experiment a: Aerosol Smoke and Smoke Water Effects on the Seedbank   In each study site, an area encompassing 30   30 m wasselected for in situ control, smoke water, and aerosolsmoke treatments. At the post-mined site, the area wascovered with topsoil that had been freshly harvestedfrom another area of the woodland. The topsoil was ap-plied to a depth of 10 cm, which mimicked the depth of collection. At the woodland site, the vegetation wascleared and the litter layer was carefully removed byhand so as not to disturb the soil seedbank. For eachtreatment, three replicate samples measuring 5   1 mwere selected at random and pegged out in each site,with a buffer zone of 1 m between all treatment repli-cates.All treatments were carried out during May (au-tumn) 1997 before the onset of winter rains. The smokewater treatments consisted of two commercially avail-able concentrated smoke water products, DC10 (com-mercially known as “Kings Park smoke water,” KingsPark and Botanic Garden, Australia) and SC63 (commer-cially known as “Regen 2000,” Grayson Trading, Austra-lia), each applied at three volumes (100, 50, and 10 mL/m  2  ) using a hand-held boom spray. The Kings Parksmoke water was produced by drawing smoke from a60-L steel combustion drum through 100 L of water for 1 hr. The aerosol smoke was applied to the soil surfaces by erecting 5   1   1-m high plastic tents over the repli-cates and fan forcing smoke from a 60-L steel combus-tion drum through a cooling pipe system into threetents simultaneously for 1 hr (Roche et al. 1997) (Fig. 2).Dry Banksia  woodland plant material was used as fuelfor the generation of smoke. After treatment the treatedsoil surface was stained brown. Including the control,there was a total of eight treatments per site.Seedling recruitment of all plant species was re-corded during November (late spring) 1997. Results foreach species were analyzed by a one-way analysis of vari-ance using the eight treatments as independent variablesto determine smoke water and aerosol smoke effects onseedling recruitment. Fisher’s protected least significantdifference at  p     0.05 was used to determine post-hoc sig-nificance.  Experiment b: Aerosol Smoke Treatment of Broadcast Seeds  A blended seed mix of 13 species routinely used in res-toration and indigenous to Banksia  woodland was usedin these trials. Species investigated were the canopy species   Allocasuarina fraseriana (  Fraser’s sheoak),   A. humilis (  dwarf sheoak),  Banksia attenuata  , and  B. menziesii  and the under-storey species  Acacia pulchella (  prickly Moses),  Beaufortia el-egans (  Myrtaceae),  Bossiaea eriocarpa (  common brown pea),  Gompholobium tomentosum (  hairy yellow pea),   Hovea pun- gens (  devil’s pins),   Hovea trisperma (  common hovea),   Jackso-nia densiflora (  Fabaceae),  Nemcia capitata  (Fabaceae), and  Xanthorrhoea preissii  (grasstree).For the ex situ trial seeds in seedling trays lined withWettex perforated cloth were randomly placed in a tentreceiving aerosol smoke for 1 hr (see above). Once Figure 2.Smoke-filled plastic tents used to treat freshly replaced topsoil in a post-mined site at Rocla Quarry Products, West-ern Australia.  Smoke, Mulch, and Seeding Effects on Woodland Restoration  188  Restoration Ecology  JUNE   2002  smoke treated, seeds were sown in plastic seedling pun-nets filled with white sand to a depth of 0.5 cm belowthe rim of the punnet. Four replicates of 100 seeds foreach species were randomly assigned to each of thepunnets for smoke-treated and untreated (control) seeds.Once spread, a very thin layer of the white sand wassieved over the seeds to stabilize the surface and seeds.Punnets were randomly positioned in a glasshouse andwatered as required. The emergence of seedlings wasrecorded after 16 weeks. Percentage germination valuesfor each species were analyzed by a t  test.In April (mid-autumn) 1996 an in situ trial was thenconducted on seeds that were nonresponsive to thesmoke treatment in the ex situ experiment. The non–smoke-responsive seed enabled the benefit of smoke asa predation deterrent in the field to be determined. Forthis trial aerosol smoke-treated seeds were sown di-rectly into a 10   10-m quadrat in a restored site spreadwith 10 cm of topsoil. The seed was sown at a rate of 4kg/ha (double the current seeding rate to facilitate scor-ing of germinants). A control plot of the same size wasseparated by a 2-m buffer strip and sown with theequivalent amount of unsmoked seed (see Roche et al.1997 for a similar design). Three replicates of 5   1 mfor each treatment were pegged, and germinants werescored in November (late spring) 1996. Germination val-ues for each species were analyzed by a t  test.  Experiment c: Mulch and Seed Broadcasting Effects  To determine the optimum thickness of mulching andthe order of mulching and broadcast seeding, trialswere conducted in subplots in areas undergoing resto-ration. The study area encompassed 0.6 ha, with the ex-perimental design consisting of eight treatments overtopsoil: control, seed mix, thin mulch (5 mm), thinmulch   seed mix under, thin mulch   seed mix over,thick mulch (10 mm), thick mulch   seed mix under,and thick mulch   seed mix over. The seed mix con-tained the canopy species B. attenuata, B. menziesii  ,   and  E. todtiana  . The understorey species in the seed mixwere   Acacia pulchella  ,   Allocasuarina fraseriana  ,   A. humilis  ,  Bossiaea eriocarpa  ,  Eremaea pauciflora (  Myrtaceae),   Hoveatrisperma  ,   Hypocalymma robustum  (Swan River myrtle),   Jacksonia densiflora  ,   J. furcellata (  Fabaceae),  Kunzea ericifolia(  spearwood),   Melaleuca scabra  (rough honeymyrtle),   M.trichopylla (  Myrtaceae), and Regelia inops (  Myrtaceae). Amulch of the canopy material from the adjacent nativevegetation was made using a large-scale mulcher.All treatments were carried out during May 1997 be-fore the onset of winter rains with three 5   1-m repli-cate plots pegged in each of the eight treatment plots.Seedling recruitment of all plant species was recordedduring the first week in November 1997. Results foreach species were analyzed by a one-way analysis of variance using the eight treatments as independentvariables. Fisher’s protected least significant differenceat  p     0.05 was used to determine post-hoc significance.  Results  Experiment a: Aerosol Smoke and Smoke Water Effects on the Seedbank   In general, the recruitment in the woodland consistedof 22 species in comparison with the 41 species re-corded in the restored site. Twenty-five species that re-cruited in the restored site did not respond to any of thegiven treatments in the woodland (Table 1). Similarly,six species that recruited in the woodland did not re-spond to any of the given treatments in the restored site(Table 2). All treatments, including the control, had higherlevels of recruitment in the restored site.Aerosol smoke significantly increased total seedlingrecruitment by 42.3-fold (Fig. 3a) and species richness by 2.8-fold (Fig. 3b) in the woodland compared with thecontrol. Nine species showed significantly higher re-cruitment with aerosol smoke treatment compared withthe control, representing 41% of the total number of spe-cies (Table 1). Another seven species showed nonsignifi-cant higher recruitment with aerosol smoke treatment.Six species that did not show a response to aerosol smokeoccurred in low densities in both aerosol smoke and con-trol plots. No species showed a negative effect as a resultof aerosol smoke application.In the restored site aerosol smoke significantly in-creased total seedling recruitment by 3.6-fold (Fig. 3a)and species richness by 1.4-fold (Fig. 3b). The lower re-sponse of this treatment, compared with the woodland,was associated with the higher level of recruitment in thecontrol. Fourteen species showed significantly higher re-cruitment with aerosol smoke treatment compared withthe control, representing over one third of the total num- ber of species (Table 2). Another 13 species showed non-significant higher recruitment with aerosol smoke treat-ment. Three species (   Adenathos cygnorum [woolly bush],   Anigozanthos manglesii [Mangle’s kangaroo paw], and  Bo-ronia ramosa  [Rutaceae]) showed a significant negative re-sponse to aerosol smoke treatment. Eleven species did notshow a response to aerosol smoke compared with control.The liquid-based smoke agents increased recruitmentat all rates in the woodland compared with the control by at least twofold (Fig. 3a). However, the only signifi-cant response (DC10 at 50 mL/m  2  ) was 23% of thatachieved with aerosol smoking. Most species showedno significant smoke water effect (Table 1). In the re-stored site, all but one treatment (DC10 at the lowestrate) increased recruitment compared with the control,with most treatments demonstrating only a 1.2-fold in-   Smoke, Mulch, and Seeding Effects on Woodland Restoration  JUNE   2002  Restoration Ecology  189  crease (Fig. 3a). However, the only significant response(DC10 at the highest rate) was 50% of that achievedwith aerosol smoking. The differences in total recruit-ment response between control and smoke water plotsand between smoke water and aerosol smoke plots inthe restored sites were not as great as those recorded inthe woodland. Again, most species showed no signifi-cant smoke water effect (Table 2). At both sites, applica-tion of smoke water failed to increase the number of species in comparison with the control (Fig. 3b). Higherlevels of recruitment (Fig. 3a) and species richness (Fig.3b) were recorded with the higher applications of thetwo smoke water agents. For example, in the restoredsite,  Astroloma macrocalyx (  swan berry),  Leucopogon spp.  (  Epacridaceae),  Nemcia capitata (  Fabaceae),  Scholtzia in-volucrata (  spiked scholtzia),  Stylidium spp.  (  Stylidiaceae),and  Verticordia nitens  (Morrison featherflower) showedsignificant effects of increasing rates of the smoke wateragents (Table 2).  Experiment b: Aerosol Smoke Treatment of Broadcast Seeds  The application of smoke ex situ to a range of speciesdid not have an effect on nine species, including six of the seven legumes, B. elegans (  Myrtaceae),  Allocasuarina fraseriana  ,   and  Allocasuarina humilis  (Fig. 4). The applica-tion of smoke to these broadcast seeds did not substan-tially alter total in situ seedling recruitment (i.e., topsoiland broadcast species), with 127   21.6 versus 143  35.6 seedlings per 5 m 2 in the control. The application of smoke to the nine broadcast seeds that were unrespon-sive to smoke ex situ did not alter their seedling recruit-ment levels. However, four species present only as top-soil stored seed showed a significant positive germinationresponse after application of smoked broadcast seed tothe site: Scholtzia involucrata  (0.3   0.3 vs. 3.3   0.9 seed-lings per 5 m 2 ,  p     0.0335),  Hibbertia huegelii  (Dilleni-aceae) (0 vs. 1.3   0.3 seedlings per 5 m 2 ,  p     0.0161),  H.subvaginata  (Dilleniaceae) (0.7   0.7 vs. 4.0   0.6 seedlingsper 5 m 2 ,  p     0.0194), and Scaevola paludosa  (Goodeni-aceae) (0 vs. 1.3   0.3 seedlings per 5 m 2 ,  p     0.0161). Experiment c: Effect of Mulch and Broadcast Seeds In general, mulch not only inhibited the recruitment of topsoil and broadcast species, but applying broadcastseeds after mulch application substantially limited re-cruitment. Seed broadcasting over topsoil contributed36% total seedling recruitment compared with control(Fig. 5a) and an additional seven species (Fig. 5b). Of the 16 species broadcast, only 11 species showed a re-sponse in at least one of the six treatments involving broadcast seeding. Mulch did not contribute any ger-minable bradysporous species. Furthermore, applyingmulch over topsoil significantly decreased seedling re-cruitment and species richness by up to 64 and 50%, re- Table 1. Aerosol smoke and smoke water effects on seedling recruitment of Banksia  woodland species in Banksia  woodland. Smoke WaterSpecies representedControlDC10 (10 mL/m 2 )DC10 (50 mL/m 2 )DC10 (100 mL/m 2 )SC63 (10 mL/m 2 )SC63 (50 mL/m 2 )SC63 (100 mL/m 2 )Aerosol Smoke  Acacia huegelii 000000.33   0.3300.33   0.33  Astroloma macrocalyx 0.33   0.33000.33   0.330000 Banksia attenuata 0 a 0 a 0.33   0.33 ab  0 a 0.33   0.33 ab  0 a 0.33   0.33 ab  1   0.58  b Banksia menziesii 0.67   0.670000000 Boronia ramosa 0 a 0 a 0 a 0 a 0 a 0 a 0 a 1.33   0.33  b Bossiaea eriocarpa 0.33   0.33 ab 0 a 0 a 0 a 0 a 0.67   0.33  b 0 a 1.33   0.67  b Calytrix sp.0.67   0.33 ab 0 a 1   0.58 ab 4.67   3.18  b 1.33   0.88 ab 0 a 3.67   3.67 ab 0.33   0.33 ab Conospermum stoechadis 0.67   0.33 ab 0 a 0 a 0 a 0 a 0.33   0.33 ab  0 a 4.67   3.71  b Conostylis aculeata 0 a 0.33   0.33 a 0.33   0.33 a 0.67   0.67  a 0.33   0.33 a 0.67   0.33 a 0.67   0.67 a 16   5.69  b Dampiera linearis 000.33   0.330000.67   0.332   2 Gompholobium tomentosum 1   0.58 ab 1   0.58 ab 0.67   0.33 ab 1   0.58 ab 0.33   0.33 a 3   1.53  b 2.33   1.45  b 1.67   0.67  b Gonocarpus pithyoides 000.33   0.3300000  Hibbertia racemosa 1.67   0.33 a 7.33   2.73 ac 14.3   11.3 ac 7   3.79 ac 6.67   3.53 ac 17   6.66 ac 22.3   13.6 c 133   32.9  b  Hibbertia subvaginata 0 a 0 a 0 a 0 a 0.33   0.33 ac 0.33   0.33 ac 0.67   0.33 c 1.67   0.33  b Laxmannia spp.0 a 0 a 1.67   1.2 a 2.33   1.86 a 0.33   0.33 a 0.67   0.67 a 0.33   0.33 a 17   8.33  b Leucopogon spp.0 a 1.33   0.88 a 6   5.03 a 10.67   9.7 ab 0.67   0.67 a 6.33   2.85 a 6   3.06 a 32.7   21.3  b Nemcia capitata 00000000.67   0.67 Opercularia vaginata 0 a 0 a 0 a 0 a 0 a 0 a 0 a 6   3.79  b Patersonia occidentalis 0000.33   0.330000 Scholtzia involucrata 01   102.33   1.8602.67   2.190.33   0.330 Stylidium spp.0.33   0.33 a 2   1.15 ac 32.7   25.7  b 11   4.16 ab 13.33   4.9  bc 6.67   3.18 ab 3.67   0.33 ab 21.3   13.8  b Thysanotus dichotomus 0000.33   0.330000.33   0.33 Values are seedlings per 5 m 2     SE. Different letters indicate significant differences using Fishers PLSD  p   0.05.Species that did not respond to any of the given treatments : Adenanthos cygnorum, Anigozanthos manglesii, Arnocrinum preissii, Beaufortia elegans, Comesperma calymega,Conostylis candicans, C. setigera, Desmocladus flexuosus, Eremaea pauciflora, Hibbertia huegelii, H. hypericoides, Hovea pungens, H. trisperma, Jacksonia densiflora, Leschenaultia floribunda, Lomandra spp., Lyginia imberbis, Macarthuria australis, Melalueca huegeleii, Stirlingia latifolia, Thysanotus arenarius, T. manglesianus, Unknown grass 1, and  Verti-cordia nitens.
Similar documents
View more...
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