Use fences or barriers to protect freshwater wetlands planted with trees/shrubs

Supporting evidence from individual studies

1. A replicated, paired, controlled study in 1985–1987 in a floodplain swamp in Louisiana, USA (Conner & Flynn 1989) reported that using chickenwire fencing to exclude herbivores increased survival of baldcypress Taxodium distichum seedlings planted in spring, but had no clear effect on survival of seedlings planted in autumn. Statistical significance was not assessed. For seedlings planted in spring, those surrounded by fencing had an 81–91% survival rate after one growing season and a 40–70% survival rate after three growing seasons. Unfenced spring-planted seedlings were all eaten within the first growing season. For seedlings planted in autumn, there was no clear effect of fencing on overall survival rates (fenced: 20–88%; unfenced: 24–68% after two growing seasons). Methods: Through 1985, three plots were each planted with 250 baldcypress seedlings: 200 in February/March and 50 in September. Of these seedlings, 75 were protected with chickenwire fencing whilst 175 were left unfenced (open to herbivory by nutria Myocastor coypus). Seedlings were stored cold (4^°C) before planting. Plots contained other trees (330–590 stems/ha) and saplings/shrubs (1,000–3,500 stems/ha) and had variable water levels. Survival of planted baldcypress seedlings was recorded in October 1985–1987.

   Study and other actions tested

   Referenced paper

   Conner W.H. & Flynn K. (1989) Growth and survival of baldcypress (Taxodium distichum[L.] Rich.) planted across a flooding gradient in a Louisiana bottomland forest. Wetlands, 9, 207-217.

2. A replicated, randomized, controlled study in 1992 in a freshwater marsh in Louisiana, USA (Myers et al. 1995) found that planted baldcypress Taxodium distichum seedlings protected with plastic guards or a sticky, insect-repellent oil grew more than unprotected seedlings. Over one growing season, seedlings within plastic guards grew significantly more than unguarded seedlings in both height (data not reported) and diameter (guarded: 0.73–0.85 cm; unguarded: 0.28–0.32 cm). Amongst guarded seedlings, growth was similar whether the guards were PVC tubes (0.74 cm diameter increase) or commercial Tubex guards (0.73–0.85 cm diameter increase). Amongst unguarded seedlings, those painted with sticky oil grew significantly more than unpainted seedlings in height (data not reported) but not diameter (painted: 0.32; unpainted: 0.28 cm diameter increase). Painted seedlings grew less than guarded seedlings in both diameter and height. Methods: In January 1992, four hundred baldcypress seedlings were planted into a marsh – with the aim of restoring the swamp that was logged around 80 years previously. Guards against nutria Myocastor coypus herbivory were placed around 240 random seedlings (80 PVC tubing, 80 beige Tubex, 80 white Tubex). Sticky oil was painted onto the lower third of 80 random seedlings. The final 80 seedlings received no protection. Some protected and unprotected seedlings received additional treatments: fertilization or removal of competing vines. Seedling diameter and height were measured at planting (January 1992) and, for surviving seedlings, after one growing season (October 1992).

   Study and other actions tested

   Referenced paper

   Myers R.S., Shaffer G.P. & Llewellyn D.W. (1995) Baldcypress (Taxodium distichum (L.) Rich.) restoration in southeast Louisiana: the relative effects of herbivory, flooding, competition, and macronutrients. Wetlands, 15, 141-148.

3. A replicated, paired, controlled study in 2006–2009 in a floodplain swamp clearing in Wisconsin, USA (Thomsen et al. 2012) found that fencing to exclude deer before sowing tree seeds had no significant effect on tree seedling abundance but increased seedling height. After roughly three years, fenced plots contained a statistically similar number of tree seedlings (40 seedlings/m²) to open plots (28 seedlings/m²). However, seedlings in fenced plots were significantly taller (73 cm) than those in open plots (46 cm). Methods: In November 2006, sixteen pairs of 2.25-m² plots were established in a floodplain swamp restoration site (clearing created by a storm; invasive vegetation recently removed and ground disked). In each pair, one plot was fenced to exclude white-tailed deer Odocoileus virginianus (plastic mesh fence, 2 m tall) whilst the other was left open. Seeds of five tree species were sown in and around all plots between 2006 and 2009. The plots were also treated regularly with herbicide, to control an invasive grass, between 2006 and 2008). Seedlings were counted and measured in August 2009.

   Study and other actions tested

   Referenced paper

   Thomsen M., Brownell K., Groshek M. & Kirsch E. (2012) Control of reed canarygrass promotes wetland herb and tree seedling establishment in an Upper Mississippi River floodplain forest. Wetlands, 32, 543-555.

4. A replicated, paired, controlled study in 2014–2015 in two degraded floodplain swamps in Victoria, Australia (Greet et al. 2016) reported that fencing to exclude browsing and grazing mammals increased survival of planted swamp gum Eucalyptus camphora seedlings. Over one year, seedlings within fenced plots had a 98–100% survival rate. In contrast, seedlings in unfenced plots had a 0–4% survival rate. Methods: In March 2014, swamp gum seedlings were planted into eighteen 100-m² plots across two floodplain wetlands (50 seedlings/plot). In each wetland, eight plots had been fenced and one was left open. All plots had been recently cut and sprayed with herbicide (to control reed canarygrass Phalaris arundinacea or common reed Phragmites australis), and planted with native shrubs and herbs along with swamp gum. Some fenced plots were also covered with matting or woodchips. Seedling survival was monitored in March 2015.

   Study and other actions tested

   Referenced paper

   Greet J., King E. & Stewart-Howie M. (2016) Plastic weed matting is better than jute or woodchips for controlling the invasive wetland grass species Phalaris arundinacea, but not Phragmites australis. Plant Protection Quarterly, 31, 19-22.

5. A replicated, paired, controlled study in 2008–2013 in two created freshwater swamps in Michigan, USA (Kangas et al. 2016) reported that fencing to exclude white-tailed deer Odocoileus virginianus typically had no significant effect on the survival or height of planted white cedar Thuja occidentalis. After five years, cedar survival was statistically similar in fenced and open plots in three of four comparisons (fenced: 2–61%; open: 0–54%). Between two and five years after planting, the average height of surviving trees changed by a similar amount in fenced and open plots in three of four comparisons (fenced: 2–39 cm/year increase; open: 2 cm/year decrease to 30 cm/year increase). In the other comparisons, involving trees planted on mounds in a site with high browsing pressure, fenced plots supported higher survival (fenced: 94%; open: 70%) and height increase (fenced: 23 cm/year; open: 2 cm/year). Methods: In spring 2008, one-year-old white-cedar seedlings were planted into 37 plots on two recently excavated wetlands (5–106 seedlings/plot, approximately 2.8 m apart). Twenty-one plots were surrounded by 3-m-tall deer-exclusion fencing. Sixteen plots were left open. In some fenced and open plots, seedlings were planted on mounds. Surviving trees were monitored in April 2010 and October 2013.

   Study and other actions tested

   Referenced paper

   Kangas L.C., Schwartz R., Pennington M.R., Webster C.R. & Chimner R.A. (2016) Artificial microtopography and herbivory protection facilitates wetland tree (Thuja occidentalis L.) survival and growth in created wetlands. New forests, 47, 73-86.