Page  43 ï~~2000 THE MICHIGAN BOTANIST 43 EFFECTIVENESS OF HAND-PULLING THE INVASIVE MOSSY STONECROP (SEDUM ACRE L.) FROM ALVAR PAVEMENTS Judith Jones Winter Spider Eco-Consulting, R.R. #1, Sheguiandah, Ontario POP 1WO (705) 368-0734 <j2@j2.nu> ABSTRACT The Eurasian weed mossy stonecrop (Sedum acre L.), a tiny succulent plant of the Crassulaceae, is present on some alvar pavements on Manitoulin Island, Ontario. This study examined the rate of spread of mossy stonecrop on alvar pavement, and tested whether hand removal of stonecrop would eradicate the weed or possibly cause an increase in regrowth of it or other weeds. The study also looked at regrowth of mossy stonecrop in situ from fragments. Nine 1 x 1 m plots were given three different treatments: no pulling; careless pulling leaving fragments; very careful pulling leaving no fragments. Plots were observed for three growing seasons after treatment. Results show stonecrop is capable of rapid, aggressive growth, but that this does not always occur. Stonecrops do regrow from fragments or possibly from root remnants. Careful pulling techniques resulted in almost no stonecrop regrowth after three growing seasons. No regrowth of other weed species occurred. INTRODUCTION Mossy stonecrop (Sedum acre L.), a common Eurasian weed, is a fibrousrooted, much-branched perennial with stems 5-10 cm long and succulent leaves of 2-5 mm (Cronquist 1991). The plants are mat-forming in habit, with a few centimeters at the tip of each creeping stem becoming upright and fleshy (Figure 1). The upright stems support terminal, small, yellow flowers and eventually dry capsular fruits. The plant has been cultivated in many forms and is widely naturalized throughout North America (Voss 1985). On Manitoulin Island, Ontario, mossy stonecrop is a common weed of pastures and open areas of shallow soil over flat limestone bedrock. In fact, the plant is able to grow on the surface of bare limestone in small patches of soil less than 1 cm deep (personal observation). Because it can tolerate such conditions, mossy stonecrop finds suitable habitat in rare alvar communities. Alvars are open, treeless ecosystems based on horizontal limestone bedrock. The vegetation is usually dominated by graminoid herbs or dwarf shrubs. Soil cover is shallow and sporadic, making these ecosystems fragile end easily disturbed by soil displacement. Recent work has brought attention to the rarity of different alvar community types as well as to the high incidence of rare species present on alvars (Brownell & Riley 2000; Reschke et al. 1999; Catling 1995; Catling & Brownell 1995). Efforts to protect high quality examples of alvar require long-term management strategies that include plans to cope with threats. Thus, questions have been

Page  44 ï~~44 THE MICHIGAN BOTANIST Vol. 39 44 THE MICHIGAN BOTANIST Vol. 39 FIGURE 1. Mossy stonecrop (Sedum acre L.). raised about the invasiveness of mossy stonecrop on alvar: whether it spreads rapidly, whether it excludes native plants from using a portion of the habitat, and whether the disturbance caused by its removal would increase renewed growth of stone crop or other weed species. Work on common St. John's wort (Hypericum perforatum L.) by Stephenson (1995) in the alvar grassland of Maxton Plains, Drummond Island, Michigan, showed that hand pulling was ineffective for that species since plants from seeds stored in the soil more than replaced pulled plants. There were additional concerns about pulling mossy stonecrop that didn't occur with other weeds in other situations. For instance, mossy stonecrop roots are thread-like, can be >10 cm long, and are difficult to remove in their entirety because they break. In addition, because the plants grow in areas of such shallow soil, it was thought that hand pulling to remove them might do more damage from soil displacement than good. As well, like most members of the Crassulaceae, mossy stonecrop fragments very readily and easily regrows from both leaf and stem fragments (personal observation). Thus, hand pulling might accidentally increase plant dispersal and regrowth. The selective use of herbicide was not considered as a control method because stonecrop occurs in small, shallow pockets of soil sitting on impervious bedrock. This and the small stature of the weed would make it nearly impossible to prevent the herbicide from pooling on the surface of the rock and affecting other plants.

Page  45 ï~~2000 THE MICHIGAN BOTANIST 45 2000 THE MICHIGAN BOTANIST 45 No r t h C h a n n e l 00 MiseryBay Provincial Nature Reserve N L a 10 km yV FIGURE 2. Location of Manitoulin Island and Misery Bay Provincial Nature Reserve. Misery Bay Provincial Nature Reserve (450 47' N 820 45' W; Figure 2) is an 880 hectare provincial park which includes approximately 80 hectares of alvar vegetation (Jones 2000). Mossy stonecrop is present in several places in a variety of alvar vegetation types at Misery Bay. Park management expressed concern about the spread of mossy stonecrop within park alvars as early as 1994 when initial park planning documents were being drafted (Will Kershaw, Ontario Parks, pers. comm.), but an action plan on the weed's removal was delayed until the effects of pulling it could be studied. This current study looked at the degree of spread of mossy stonecrop on the alvar at Misery Bay Provincial Nature Reserve, its regeneration from fragments, and the effects of hand pulling on the population of stonecrop and other weeds. METHODS Nine permanent 1 x 1 m plots were laid out in an area of alvar pavement near the Lake Huron shore. To make the plots inconspicuous, only the corners were flagged. During monitoring, meter sticks were placed along the 4 sides of the plots. Before any weed removal was done, each plot was sketched and photographed, and an estimate of the percent cover of stonecrop present in each plot was made using the meter sticks as a scale. The number of clumps of stonecrop was also counted. Mossy stonecrop was then pulled from the area im around the outside of all plots, to prevent any immediate reseeding or recolonization from fragments outside plot boundaries. Three sets of three replicate plots were created. Each set was given a different treatment. The first treatment ("C plots") was a control in which no pulling or other action was done. The second treatment ("F plots") was hand pulling in a careless fashion, such as might be done by an unconcerned worker, intentionally leaving behind a small number of fragments at documented points. The third

Page  46 ï~~46 THE MICHIGAN BOTANIST Vol. 39 treatment ("E plots") was hand pulling with extreme care leaving behind no known fragments in the best attempt at eradication. The technique was such that the fingers surrounded the clumps and reached under the plant so that most fragments remained in the hand while the other hand was held underneath the first hand to catch anything which fell out. All removed plants were carefully placed in plastic bags, removed from the park and composted. F and E Plots were photographed again following stonecrop removal. Set-up took place in early July, 1998, with some plants in flower or young developing fruit but before fruit maturity. Soil was powdery dry although the plants appeared turgid. No other weed species was observed in the area. The first monitoring took place September 18, 1999 (a little more than one year after the set-up of the experiment) after two growing seasons. Monitoring consisted of estimating the percent cover of mossy stonecrop, counting the number of clumps of it in the plot, sketching the location of the clumps in the plot, and photographing the plot. Notes were also made of any presence of other weeds in the plot. A second monitoring took place August 21, 2000 (a little over two years after the initial set-up) after three growing seasons. The weed-free 1 m zone around all plots was maintained at each monitoring. It is intended that this study will continue for several more years. RESULTS Table 1 shows some of the changes in the 9 plots over three growing seasons. Control plots showed little net increase in percent cover or number of plants in three growing seasons. The number of plants in C2 increased with a high number of new single stems, but these are small in size and thus did not result in an increase of percent cover of stonecrop. Similarly, in C3 an increase in the number of tiny new single stems was off-set by a major die-back of larger clumps resulting in a net reduction in percent cover of stonecrop. In fact, the overall condition of the control plots after three seasons of growth was a major die-back of large clumps with only a few areas supporting new, tiny, single stems. The first fragmentation plot, Fl, showed no renewal of mossy stonecrop at all during the course of the three growing seasons, and no other weeds became present. The second fragmentation plot, F2, had two small clumps of mossy stonecrop present at the second monitoring, both were present at locations where fragments had been left. The clumps were small, one having 2 branches, the other with 9 branches, probably indicating one year's growth. There was an overall net loss of percent cover and number of clumps of stonecrop. These clumps were removed using the most careful technique, and the F2 plot was completely stonecrop-free at the end of the second growing season in 2000 (Figure 3). Fragmentation plot F3 had no stonecrop present at the end of the second growing season. However, at the end of the third growing season, in 2000, one clump of mossy stonecrop was present beside and reaching underneath a large boulder. This was a location where fragments had been left after the initial treatment in 1998. This clump was removed in 2000 with the most careful technique. Eradication plots El and E3 showed no renewed presence of mossy stonecrop at the end of 1999 or 2000. Plot E2 had one new, tiny stem present at the end of 1999 growing out of a dense patch of creeping juniper (Juniperus horizontalis

Page  47 ï~~2000 THE MICHIGAN BOTANIST 47 TABLE 1. Changes in nine plots over three growing seasons. C plots are controls which had no pulling. F plots had some fragments left behind in the first treatment. E plots had complete eradication attempted. Any stonecrop found in 1999 in F or E plots was removed using the most careful pulling techniques. number of clumps percent cover Plot name initial 1999 2000 initial 1999 2000 C1 25 -25 24 5% 5% 3% C2 33 41 40 7% 7% 8% C3 5 21 mostly 15 5% 3% 4% tiny new stems F1 4 0 0 0 0 0 F2 6 2 new stems 0 0 <1% 0 F3 2 0 1 0 0 <1% El 3 0 0 0 0 0 E2 5 1 0 0 <1% 0 E3 2 0 0 0 0 0 Moench). This stem was removed with the most careful technique. At the end of 2000, all E plots were completely free of mossy stonecrop. As well, no new presence of any other weed species occurred. On the other hand, new colonization by the native species rock sandwort (Minuartia michauxii (Fenzl)Farw.) was noted at three spots in plot F2 in 2000. DISCUSSION In fragmentation plot F3, it is unclear how the regrowth of mossy stonecrop occurred. The location of the clump coincides with fragments left at the initial pulling, making it highly likely that the re-growth occurred from fragments, but it is puzzling why growth was seen at this location in the third growing season but not in the second. It is not known if the fragments survived a season without growth, whether they were inconspicuous in the first year, or whether the clump regrew from roots remaining from the initial pulled plants. This requires further study. The tiny stem present in the creeping juniper (Juniperus horizontalis Moench) in plot E2 is probably from a fragment that was overlooked in the first treatment because it fell into the crevices of the shrub. Still, it could also be from a root fragment or from outside dispersal. Observation of the three control plots for three growing seasons portrays the degree of spread of mossy stonecrop on alvar to be a very slow process, interrupted by major die-backs, and which could hardly be termed aggressive or invasive. However, the explosion of a large (300 cm2), completely new patch of stonecrop in the removal zone adjacent to control plot C2 (Figure 4) demonstrates that the species is capable of rapid growth. Observations on the rate of spread of mossy stonecrop remain inconclusive, but the observed potential for occasional rapid growth remains a concern. Further work is needed to shed light on the conditions which allow such rapid and potentially invasive growth.

Page  48 ï~~48 THE MICHIGAN BOTANIST Vol. 39 48 THE MICHIGAN BOTANIST Vol. 39 FIGURE 3. Plot F1 before pulling (above) and after three growing seasons in 2000 (below). No mossy stonecrop is present in 2000.

Page  49 ï~~2000 THE MICHIGAN BOTANIST 49 Results from the fragmentation plots show that it is highly probable that mossy stonecrop can re-grow from fragments, although there is a small possibility that the re-growth could have been from root remnants unknowingly not removed. Yet, from a pragmatic point of view, even with the re-growth, the net result is a reduction in the percent cover and number of clumps of mossy stonecrop. This means that even if alvar managers must rely on careless volunteers to pull the stonecrop and are then required to go out and pull again for a second season or perhaps additional seasons, the subsequent efforts should require much less time and effort. The initial results of this study show that in spite of the small number of renewals of mossy stonecrop, the net result of hand pulling is a major reduction in the presence of the weed. Results from the eradication plots show that careful pulling can very nearly eliminate mossy stonecrop, so ideally workers pulling the plants should use careful technique and vigilantly remove any fragments that get dropped. None of the plots showed any new growth of other weed species, but the location of the plots is virtually free of other weed species. Therefore, this experiment is not a good test of the effects of soil disturbance from pulling stonecrop on the regrowth of other weeds. The new growth of the native rock sandwort in plot F2 could perhaps point to an example of stonecrop filling a niche which would normally be occupied by a native alvar species. It is possible that soil disturbance from weed removal may even have helped the rock sandwort get established. N iv ".:........\ \ vP\,~ FIGURE 4. A very large patch of mossy stonecrop (adjacent to Plot C2) which grew to this size in a single year. The rapid growth of this patch suggests the species is capable of aggressive spread, despite the fact that this was not demonstrated in the control plots.

Page  50 ï~~50 THE MICHIGAN BOTANIST Vol. 39 CONCLUSIONS The initial results of this long-term study show that mossy stonecrop may be hand pulled from alvar habitats with little regrowth. It is suggested that limited hand pulling be initiated in a wider but controlled area which will be monitored. The potential for trampling of the habitat by workers must be addressed, and care should be taken as it is not known if walking on mossy stonecrop could affect dispersal or re-growth. ACKNOWLEDGMENTS Thanks to the Friends of Misery Bay and Ontario Parks for providing access to the study site at Misery Bay Provincial Nature Reserve. Thanks to Steve Hall for field assistance. REFERENCES Brownell, V.R. & J.L. Riley, 2000 The Alvars of Ontario. Federation of Ontario Naturalists, Don Mills, Ontario. 269 pp. Catling, P.M., 1995. The extent of confinement of vascular plants to alvars in Southern Ontario. Canadian Field Naturalist 109:172-181. Catling, P.M. & Brownell, V.R., 1995. A review of the alvars of the Great Lakes Region: distribution, floristic composition, biogeography and protection. Canadian Field Naturalist 109:143-171. Cronquist, A., 1991. Manual of Vascular Plants of Northeastern United States and Adjacent Canada. 2nd ed. New York Botanical Garden, Bronx, NY. 910 pp. Jones, J., 2000. Manitoulin coast and uplands study area: Blue Jay Creek Provincial Park, Mac's Bay Conservation Reserve, Misery Bay Provincial Nature Reserve, Queen Elizabeth Queen Mother M'Nidoo M'Nissing ("Spirit Island") Proposed Provincial Park. Unpublished report to Ontario Parks, Ontario Ministry of Natural Resources, Sudbury, Ontario. 151 pp. Reschke, C., R. Reid, J. Jones, T. Feeney, & H. Potter, 1999. Conserving Great Lakes Alvars: Final technical report of the International Alvar Conservation Initiative. The Nature Conservancy, Great Lakes Program Office, Chicago, IL. 230 pp. Stephenson, S.N., 1995. Annual report to The Nature Conservancy on long-term monitoring activities on the Maxton Plains Preserve. 23 pp. Voss, E.G., 1985. Michigan Flora, part II Dicots. Cranbrook Institute of Science Bulletin 59 and The University of Michigan Herbarium., Ann Arbor, MI. 724 pp.