Evaluation of alternative approaches to rainforest restoration on abandoned pasturelands in tropical North Queensland,Australia

The lag time for natural recruitment of tropical rainforest species in abandoned pastureland is very long, therefore artificial restoration techniques have been employed to accelerate natural seedling recruitment. The objectives of this study were to investigate: (1) the success/failure of establishment 502 seedlings belonging to 15 species from 11 families planted approximately ten years ago; and (2) the influence of different restoration techniques on enhancing natural recruitment during this period. The study was conducted in the wet tropical rainforest region of northeast Queensland, Australia as a completely randomized block design involving five treatments with two replicates. In each plot, 63 tropical rainforest seedlings from one or a combination of species were planted randomly. Two control plots were laid out where no seedlings were planted. Survival, height and diameter data were taken on the seedlings ten years after planting. Each 11×17 m² plot was further divided into 187, 1×1 m² subplots. Within each subplot all seedlings recruited were located and identified. Canopy cover was estimated using belt transects 1 m apart that ran in an east–west direction across the plots. Within each plot the percentage of grass, and the crown cover were estimated using the Braun‐Blanquet cover abundance scale. Survival rate of planted seedlings varied across the treatment plots. The survival rate ranged from 65 to 75 per cent for primary‐promoter species, 85 to 100 per cent in middle‐phase species and 42 to 57 per cent for mature‐phase species. No Pilidiostigma tropicum seedlings survived in any treatment. Fourteen species recruited naturally across the treatment plots. A total of 410 seedlings were naturally recruited from 11 different families in the ten‐year‐old reforested site. The highest natural recruitment (236 seedlings) occurred in Treatment 3, where Omalanthus novo‐guineensis seedlings were planted with eight primary‐promoter species, followed by 99 in Treatment 5 where a group of primary‐promoters, middle phase species and mature‐phase species were planted together, 36 in Treatment 4 (Alphitonia petriei planted with eight primary‐promoter species), 10 in Treatment 2 where only Omalanthus novo‐guineensis seedlings were planted, and 13 in control plots. Grass cover declined with increasing species diversity and increased canopy cover. The results indicate that the diversity of species used in restoration had a major influence on natural recruitment. Copyright © 2004 John Wiley & Sons, Ltd.