Sow grassland seeds from a local source

Supporting evidence from individual studies

1. A replicated, controlled study in 2004–2010 at a former mining site in Saxony-Anhalt, Germany (Kirmer et al. 2012) found that sowing grassland seeds from a local source, along with increasing the number of species in a seed mix, led to an increase in the species richness and cover of target plants. After six years, plots sown with a high diversity local seed mix had on average a greater number and cover of target plant species (28 species, 83% cover) than plots sown with a low diversity non-local seed mix (12 species, 36% cover). In December 2004, three blocks were established on an unvegetated area (240 x 50 m) of boulder clay mixed with sand. In each block, one plot was sown with a high diversity mix of seeds from a local source (11 grass and 40 herb species, sown at 36 kg/ha), and one plot was sown with a low diversity mix of non-local seeds (three grass cultivars, sown at 100 kg/ha). Vegetation was recorded annually within a 5-m² quadrat in each plot in 2005–2010.

   Study and other actions tested

   Referenced paper

   Kirmer A., Baasch A. & Tischew S. (2012) Sowing of low and high diversity seed mixtures in ecological restoration of surface mined-land. Applied Vegetation Science, 15, 198-207.

2. A replicated, site comparison study in 2005–2007 in 15 former arable fields in Colorado, USA (Munson & Lauenroth 2014) found that sowing grass and forb seeds from a local source had no effect on vegetation height and mixed effects on vegetation cover and plant density compared to sowing non-native grass seeds. In areas where local grass and forb seeds were sown 18–20 years previously, plant cover was higher (20–24%) than in areas sown with non-native grass seeds (12–19%) but there was no significant difference for areas where local seeds had been sown 2–9 years ago (8–12%). There was no significant difference in plant height between areas where local seeds were sown (14–17 cm) and areas where non-native seeds were sown (15–31 cm). Plant density was higher in areas where local seeds were sown 2–9 years ago than in areas where non-native seeds were sown (18–61 plants/m² vs 17–43 plants/m²), but plant density in areas where local seeds were sown 18–20 years ago (18–34 plants/m²) did not significantly differ from areas where non-native seeds were sown. Each field was sown with either a mix of perennial grass and forb seeds of species native to northern Colorado (nine fields), or with seeds of the non-native grasses, wheatgrass Agropyron intermedium (three fields) or smooth brome Bromus inermis (three fields). In August–September 2005–2007, forty 0.25^-m² circular plots were placed in each field and plant height, density and cover were estimated.

   Study and other actions tested

   Referenced paper

   Munson S. & Lauenroth W.K. (2014) Controls of vegetation structure and net primary production in restored grasslands. Journal of Applied Ecology, 51, 988-996.

3. A replicated, randomized, controlled experiment in 2012 in a greenhouse in Wyoming, USA (Herget et al. 2015) found that sowing Sandberg bluegrass Poa secunda seeds from a local source resulted in similar survival to that found when non-local varieties were sown. After four months, the survival of Sandberg bluegrass plants from locally sourced seeds (97–98%) was not significantly different from that of plants from non-local seeds (93–100%). In December 2012, wild Sandberg bluegrass seeds sourced from two local sites were sown in 80 pots, while cultivated Sandberg bluegrass seeds from three sources were sown in 120 pots. All pots were filled with sand and peat moss, and watered twice/day. Two weeks after planting, emerged seedlings were thinned to one plant/pot. Plant survival was monitored for four months from December to March 2012.

   Study and other actions tested

   Referenced paper

   Herget M.E., Hufford K.M., Mummey D.L., Mealor B.A. & Shreading L.N. (2015) Effects of competition with Bromus tectorum on early establishment of Poa secunda accessions: can seed source impact restoration success? Restoration Ecology, 23, 277-283.

4. A replicated, randomized, paired, controlled study in 2011–2013 in a former plant nursery in southern Germany (Walker et al. 2015) found that sowing locally sourced seeds reduced the abundance of the invasive plant fleabane Erigeron annuus, had a mixed effect on the abundance of individual plant species but did not alter plant cover. There were more Carthusian pink Dianthus carthusianorum plants in areas where a local grass seeds were sown (56–96 plants/m²) than in areas where commercial grass seeds were sown (31–58 plants/m²). However, there was no significant difference in the abundance of perennial flax Linum perenne (local: 84–109 plants/m²; commercial: 72–122 plants/m²) and ox-eye Buphthalmum salicifolium (local: 22–38 plants/m²; commercial: 24–40 plants/m²). Vegetation cover did not differ significantly between areas where seeds were sown (43–50%) and where no seeds were sown (34%). There were fewer invasive annual fleabane plants in areas where seeds were sown (108–142 plants/m²) than in areas where no seeds were sown (141 plants/m²). Before seeds were sown, the site was sprayed with glyphosate herbicide and was harrowed and raked. Twelve 3.5 x 1 m plots were sown with locally sourced seeds of the species red fescue Festuca rubra at a rate of 0.4 g/m², twelve plots were sown with commercially sourced red fescue seed, and six plots were not sown with grass seed. All plots were sown with seeds of the species perennial flax, Carthusian pink, and ox-eye. In July 2012, seedlings were counted in all plots, and in May 2013, plants of all species were counted.

   Study and other actions tested

   Referenced paper

   Walker E.A., Hermann J.M. & Kollman J. (2015) Grassland restoration by seeding: seed source and growth form matter more than density. Applied Vegetation Science, 18, 368-378.

5. A replicated, controlled study in 2011–2012 in an abandoned quarry site in northern Italy (Gilardelli et al. 2016) found that sowing grassland seeds from a local source increased plant species richness, but did not alter the height of herbaceous vegetation, compared to sowing a commercial seed mix. These results are not based on statistical analyses. After one year, areas sown with local seeds had higher average plant species richness (16 species/plot) than areas sown with a commercial seed mix (10 species/plot). The average height of herbaceous plants was similar between areas sown with local seeds (100 cm) and a commercial seed mix (93 cm). In June 2011, the site was remodelled to create two 200-m² terraced areas that were almost flat and topsoil was added. One area was sown with hay seeds harvested from a nearby grassland at a rate of 36 g/m², while the other was sown with a commercial seed mix of grassland species. Shrubs and trees were planted in both areas. Plants were watered in 2012 during dry periods. In June 2012, vegetation was surveyed in three 3 x 3 m randomly located plots in each area.

   Study and other actions tested

   Referenced paper

   Gilardelli F., Sgorbati S., Citterio S. & Gentili R. (2016) Restoring limestone quarries: hayseed, commercial seed mixture or spontaneous succession? Land Degradation and Development, 27, 316-324.