Species wood density and the location of planted seedlings drive early‐stage seedling survival during tropical forest restoration

Abstract

The success of restoration projects is known to vary widely, with outcomes relating to numerous biotic and abiotic factors. Though many studies have examined the factors associated with long‐term restoration success, few have examined which factors impact the establishment of restoration plantings. In Australia's Wet Tropics, we used a large replicated restoration experiment to assess seedling survival for 24 native rainforest species commonly used in local restoration efforts. The experiment allowed for a rigorous assessment of the effects of species functional traits, planting conditions, and landscape‐ and local‐scale biotic and abiotic factors on seedling survival. This study reports on seedling survival between three different time periods of 0–4, 4–9 and 9–31 months post planting. The probability of seedling survival was influenced by multiple factors, varying in importance over time. Across the whole study period, seedlings with high wood density and which were planted closer to intact forest consistently displayed the highest probabilities of survival. Transient factors affecting seedling survival across the three time periods included plot aspect (0–4 months only), the identity of the planter and slope (4–9 and 9–31 months). Overall, species survival did not differ between the low (6 species) and high (24 species) diversity treatments, but was significantly lower in monocultures of Flindersia brayleyana by the end of the study. We demonstrate that early‐stage seedling survival depends on species wood density and planting location. Our results support the use of species with more conservative growth strategies when limited funds are available for follow‐up plantings. High wood density species had significantly higher survival than lower wood density, early successional species typically used in rainforest restoration plantings. Synthesis and applications. Our study highlights the importance of wood density and landscape structure to the initial survival of rainforest plantings. Factors influencing seedling survival shifted over time but, most importantly, our results highlight that, when planting into abandoned pastures, it may be preferable to select species with higher wood densities to maximize survival during the crucial early stages of establishment and growth.

Seedling growth responses to species‐, neighborhood‐, and landscape‐scale effects during tropical forest restoration

“…Specifically, we show that species wood density is an important predictor for both seedling growth and survival (Charles et al 2018) and clearly demonstrates the mortalitygrowth tradeoff strategies commonly employed by tropical trees. Species selection can be difficult, because growth rates are affected by combinations of biotic and abiotic factors operating across multiple spatial scales.…”

“…This result was unexpected given the commonly observed negative relationship between seed mass and growth in both restoration plantings (Baraloto et al 2005, Mart ınez-Garza et al 2013a) and in mature forests (Osunkoya et al 1993, Poorter and Rose 2005, Poorter et al 2008). While larger seeded species commonly display higher survival rates (Baraloto et al 2005, Poorter et al 2008, with greater establishment success promoting further growth (Schupp 1995), our recent study of survival in the same experiment revealed a negative relationship between seed mass and seedling survival, though this was not significant (Charles et al 2018). This legacy effect may also be enhanced by the positive relationship between seed mass and specific root length (SRL; Westoby et al 1992), which can reduce post-planting seedling stress (Grossnickle 2005) and increase seedling growth (Wright and Westoby 1999).…”

“…Seedling survival status was recorded by visual inspection, with seedlings deemed to be dead if there was pronounced stem desiccation, no leaves, and/or the seedling could not be located (Charles et al 2018). Seedling survival status was recorded by visual inspection, with seedlings deemed to be dead if there was pronounced stem desiccation, no leaves, and/or the seedling could not be located (Charles et al 2018).…”

“…In particular, the slower growth in height for high wood density ❖ www.esajournals.org species is likely due to greater allocation of biomass to stem volume for structural support than shoot growth for height (Niklas 1994). However, species with low wood densities are known from this system and others to have low establishment rates in restoration plantings (Lamb 2011, Charles et al 2018). However, species with low wood densities are known from this system and others to have low establishment rates in restoration plantings (Lamb 2011, Charles et al 2018).…”

“…However, species with low wood densities are known from this system and others to have low establishment rates in restoration plantings (Lamb 2011, Charles et al 2018). that restoration success results from a combination of tree survival and tree growth, it is important to consider growth rate results in light of survival rates in the system over the same time period, which were very low in our experiment (Charles et al 2018). 3.…”