Barriers to forest regeneration of deforested and abandoned land in Panama

Summary

In Panama, abandoned agricultural land that formerly comprised tropical forest are often invaded by the exotic invasive grass Saccharum spontaneum, which precludes native forest regeneration. This study evaluated forest regeneration in Saccharum invaded areas and identified best management practices to enhance forest re-establishment.

Study site: The study was undertaken at Las Pavas, near the Panama Canal, 4 km south-west of Barro Colorado Nature Monument (BCNM). Prior to deforestation, the area supported tropical moist forest similar to the BCNM. Deforestation began in 1976, by 1984 most had been converted to subsistence agricultural land, but abandoned before 1990 and invaded by Saccharum spontaneum. Interspersed within the Saccharum were isolated trees, shrubs, and remnant forest patches.

Experimental design: In Saccharum-dominated areas, five sites (0.6 to 3.4 km apart) were selected, each adjacent to a forest patch. At each site, a transect was established into the Saccharum, perpendicular to the forest edge, with plots at 10, 35 and 85 m along it. Each plot was divided into three 12 × 12 m subplots and one of three treatments applied: cutting Saccharum once, cutting three times, and control (no cut). The Saccharum was cut with machetes in July 1996 (once cut) and at intervals 4 months apart in November 1996 and March 1997 (3-times cut). The subplots were further divided into 25, 2 m × 2 m squares. Tree and shrub seedlings and saplings within these were identified, tagged and mapped. Height and basal diameter were measured on four occasions: August 1996 (mid-wet season), December 1996 (end of wet season), April 1997 (late dry season) and mid-July 1997 (early wet season).

To study the effect of fire on tree regeneration, the design as above was repeated at each site 20–50 m distant from existing transects. A 15 m firebreak was created between the pair of transects. Plots on one side were burned during April 1997 (peak fire season). Tree and shrub seedlings and saplings were monitored for a year at four census intervals (two pre-fire and two post-fire). To determine any effect of remnant vegetation on natural regeneration, all shrubs, large monocots and isolated trees were mapped and their distance from each seedling measured.

Soil analyses: Soil samples were taken from the top 20 cm in each subplot 10 m from the forest patch and from the forest patch adjacent to each subplot. Soil pH and various soil chemical and nutirent parameters were analysed.

Abiotic factors: Cutting significantly decreased Saccharum live above-ground biomass, from 4,652 g/m (control) to 3,205 g/m (once cut) and 1,308 g/m (3-times cut). Cutting increased light levels recorded 0.5 m above ground (control 3% incident, once cut 15%, 3-times cut 39%). Total soil nitrogen levels were low and significantly lower in Saccharum grassland (average 0.37%) than in adjacent forest (0.52%). Higher soil phosphorus (burned 0.013 mg/g, unburned 0.004 mg/g) and copper levels were recorded in grassland after fire. No other nutrients differed significantly between grassland and forest, or between burned and unburned plots. Soil pH levels were high (e.g. unburned Saccharum 5.9) for tropical soils.

Regeneration: In total, 4,984 individuals of 80 species were recorded. Average basal diameter was 6 mm (range: 0.1 to 69.1 mm). Average height was 39.2 cm (range: 0.7 to 537.4 cm), with 59 over 2.5 m tall. The most common species were Gustavia superba (1,060 individuals), Piper marginatum (522), Cochlospermum vitifolium (400), Spondias mombin (290) and Cordia alliodora (241).

Tree and shrub density was lowest in the first census (August 1996) with 14.7 individuals/ 10 m², than subsequently (April 1997, 28.5; December 1996, 35.4; August 1997, 33.7). Survival was lowest in the December 1996–April 1997 dry season , (72%) and highest in the 1997 rainy season (April–August 1997, 83%). Recruitment was lowest in the dry season, falling from an average of 19 recruits/10 m² (56% of all individuals) in December 1996 to 3/10 m² (15% of all individuals) in April 1997. Tree seedling growth was also less during the dry season.

The number of species was highest near the forest and lowest at the intermediate distance (35 m). Relative growth rate of seedlings was highest in the 3-times cut Saccharum (average 72%/year), 34%/year (once cut) and 25%/year (uncut). No other response variables were significantly affected by distance from the forest or treatment.

Burning decreased the number of species (average 6.2 pre-fire to 2.8/10 m² post-fire). Percentage recruitment after fire decreased in 3-times cut Saccharum, and increased in the once cut and uncut Saccharum plots. Fire reduced average seedling density from 28.2 to 11.6/10 m² (excluding root sprouts). The percentage of seedlings originating as root sprouts increased in burned plots (unburned 31.3%, burned 51.8%). Plots closest to the forest patch had more wind-dispersed species.

When seed input was experimentally equalized, large-seeded shade-tolerant species had the highest germination and survival in the Saccharum control, whilst small-seeded species performed poorly. Small-seeded moderately light-demanding species naturally regenerated in the highest proportions.

Conclusions: Even with fire prevention, tree regeneration will not proceed unassisted; low seed dispersal limits regeneration of large-seeded species, small-seeded species are constrained by Saccharum competition. Shading effectively eliminates Saccharum, suggesting that planting trees to overcome dispersal limitations and to produce shade cover is perhaps the best way to eliminate Saccharum and promote forest regeneration. Costly Saccharum cutting treatments are not recommended by the authors as they did not significantly increase density, species richness or seedling/sapling survival. Cutting however, increased seedling growth, therefore might be considered for small-scale restorations.