In lieu of an abstract, here is a brief excerpt of the content:

  • Germination and Survival of Tree Seeds in a Tropical Montane Forest Restoration Study (Costa Rica)
  • Gabriel C. Sady (bio), Karen D. Holl (bio), Rebecca J. Cole (bio), and Rakan A. Zahawi (bio)

Planting tree seedlings is a common strategy to facilitate tropical forest recovery, but it is resource intensive. Planting trees in small patches or "island" as a restoration strategy (Zahawi and Augspurger 2006, Cole et al., forthcoming) imitates the nucleation process that occurs naturally over longer time scales from random dispersal events. Moreover, the island planting strategy is less expensive [End Page 121] than plantations and therefore may be more practical for restoring large areas, but there has been little study of its long-term effectiveness.

Click for larger view
View full resolution
Table 1.

Mean (± SD) tree canopy and ground cover, surface soil temperature (daily maximum and minimum), and light (photosynthetically active radiation—PAR values were averaged from 5:30 a.m. to 5:30 p.m.) in four microhabitats (Agua Buena, Costa Rica). Values with the same letter are not significantly different across microhabitats.

Several factors limit tropical forest recovery, including lack of seed dispersal and germination, competition from pasture grasses and other ruderal vegetation, and microclimatic extremes (reviewed in Holl 2002b). A large-scale, long-term restoration experiment in southern Costa Rica is testing the strategy of planting tree islands to overcome these limitations and facilitate tropical forest recovery (Cole et al., forthcoming). As part of this research, we studied seed germination and survival of three small-seeded (≤ 1 cm dia) tree species that naturally disperse into abandoned pastures during early succession (Cole et al., forthcoming) to determine which planting strategy best facilitates seedling establishment.

We conducted the study at four experimental restoration sites near Agua Buena, Coto Brus, Costa Rica (8°44'36" N, 82°58'04" W) from July 2007 through June 2008. All sites had been used for more than 18 years for agriculture, until 2004 when we established in each site three 50 m × 50 m treatment plots: control (no seedlings planted), tree plantation (entire plot planted with 313 seedlings spaced 2.8 m apart), and tree islands. The island treatment plot included two large islands (12 m × 12 m) used in this study that were each planted with 25 trees (16 timber trees and 9 softwoods). Two softwoods, poro (Erythrina poeppigiana) and guaba (Inga edulis), are nitrogen-fixers that are widely used as shade trees in coffee plantations. Mayo blanco (Vochysia guatemalensis) and amarillón (Terminalia amazonia) are slower-growing timber species used extensively for reforestation in the region. These four native or naturalized species were used in both planting treatments. See Cole et al. (forthcoming) for more experimental design details.

We collected seeds from fruiting trees in the area of three small-seeded species: Tabebuia rosea (wind-dispersed), Cassia fistula and Psidium guajava (both animal-dispersed). We planted 2,592 seeds without pretreating them on 24–29 June 2007 (early rainy season) in four microhabitats: plantation (P); control (C); interior (I) and at the edge (E) of large tree islands. Seeds were planted more than 5 m from the edge of plantation and control plots and at least 3 m from the edge in island interior plots. In each microhabitat, we planted two separate 1 m2 quadrats per species, each with 81 seeds 10 cm apart in a 9 × 9 grid. We placed seeds on top of the leaf litter to simulate natural dispersal. Seeds were also planted in a shade house near one site to determine timing of germination.

We recorded germination in the field one week after planting and monitored germination of new seeds and survival of seedlings every two weeks through August 2007. Thereafter, we monitored seed germination and seedling survival at about three-month intervals through June 2008, when we also measured seedling height.

During 6–15 March 2008 for each seed quadrat, we estimated percent cover classes of grasses, forbs, and bare ground (0%, ≥ 1%–5%, > 5%–10%, > 10%–25%, > 25%–50%, > 50%–75%, > 75%–95%, and > 95%–100%) and used midpoints for analyses. We also quantified canopy cover with a densiometer. For three days at each site February–March 2008, surface soil temperature...


Additional Information

Print ISSN
pp. 121-124
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.