Ecological Archives E086-129-A1

Christopher Baraloto, Deborah E. Goldberg, and Damien Bonal. 2005. Performance trade-offs among tropical tree seedlings in contrasting microhabitats. Ecology 86:2461–2472.

Appendix A. Plot environmental characteristics for tropical tree seedling experiments.

Six paired gap and understory plots were established in each of four soil types we described for Paracou, French Guiana. Few sharp ridges occur across the landscape, and the relatively flat plateaus can be characterized as white sand (podzol) or clay (plateau) soils. Midslope from clay plateau sites are clay soils with superficial drainage (slopes). Seasonally-inundated sites characterize areas of the lowest topography, which we have grouped here as swamp soils.

To determine the actual resource gradients represented by the plots, we measured environmental conditions within each plot in October 1999. Light availability, expressed as the proportion of radiation arriving at a site relative to that above the forest canopy (percentage of daylight), was estimated using hemispherical photos taken at each of the three corners of each plot (HemiView software, Delta-T Instruments, UK). In addition, three replicate 5 cm diameter soil cores of 0–10 cm depth were made, from which gravimetric moisture content was determined after oven-drying to constant mass at 100 °C. A 50 g soil subsample from the three bulked replicates was sieved to 2 mm and transported to the University of Michigan, where soil extractable phosphate was estimated colorimetrically using an ascorbic acid-molybdate complex in the Orthophosphate Procedure (880 nm) on an Alpkem Rapid Flow AutoAnalyzer (RFA 3550; OI Analytical, College Station, Texas, USA), following extraction of 5 g in 0.1 M NaHCO 3 as described in Olsen et al. (1954). Soil organic carbon and nitrogen were determined by gas chromatography following dry combustion at 700 °C of a separate 50 mg subsample, using an automated elemental combustion system (NC2500, CE Instruments, Milan, Italy).

Although site variation within a soil type was high, three of the four groups classified by 1.5-m core profiles differed significantly in gravimetric moisture during the dry season, as well as carbon-nitrogen ratios and phosphate levels for surface cores (Table A1). Soils from swamp forests had the highest levels of soil moisture as well as extractable phosphate. Surface cores from the white sand forest were rich in organic matter, with a trend toward the highest carbon-nitrogen ratios. White sand sites also showed a trend toward more extractable phosphate than either of the clay-type (plateau or slope) soils.

Soil properties did not differ between gap and understory sites, with consistent trends among soil types for gap and understory plots (Table A1). Although light transmission differed markedly between the understory and gap sites, a large degree of variation in light transmission was exhibited among gap sites within a soil type, presumably due to differences in gap size (Table A1). Still, the overall gradient in light transmission was similar among the four soil types (no gap × soil type interaction for light availability), though the midslope sites did receive relatively less light on average.

TABLE A1. Description of resource gradients for gap and understory sites in each of four soil types from the field experiment. Data are the means (SD) of six planting sites per light-soil type situation. Lettered superscripts indicate significant differences (P < 0.05) among groups for Bonferroni multiple comparisons following a two-way analysis of variance on log-transformed data.

Variable

Understory

Gap

F gap

F soil

F gap × soil

Slope

Podzol

Plateau

Swamp

Slope

Podzol

Plateau

Swamp

Light availability
(% Daylight)

0.83
(0.21)a

1.5
(1.0)a

1.0
(0.43)a

0.87
(0.45)a

5.6
(1.5)b

9.4
(6.4)b

9.2
(4.5)b

9.4
(4.7)b

213.6***

1.63

0.93

Soil Moisture
(m3 m-3)

11.5
(2.4)ab

8.1
(2.0)a

14.9
(2.1)b

35.8
(9.5)c

12.3
(2.4)ab

8.0
(1.3)a

14.1
(1.6)b

33.3
(5.4)c

0.01

48.9***

0.12

Soil phosphate-P §
(mg kg-1)

1.5
(0.5)a

2.9
(0.7)ab

1.5
(0.3)a

5.1
(1.9)b

1.8
(0.6)a

3.0
(0.7)ab

1.1
(0.2)a

5.0
(1.6)b

0.01

11.3***

0.50

Soil C:N §
(g g-1)

13.8
(1.9)a

14.2
(2.1)a

12.9
(0.8)a

13.0
(0.7)a

13.4
(1.3)a

14.9
(1.5)a

12.6
(0.7)a

13.2
(1.5)a

0.02

2.59

0.29

* P < 0.05; ** P < 0.01; *** P < 0.001

Light measures are the mean GSF index from 3 subplots within each planting site (n = 18), measured from hemispherical photos at 1 m from soil surface.

Soil moisture measurements are the means of 0–10 cm cores taken from the three subplots.

§ Soil extractable phosphorus (via the Olsen NaHCO 3 method) and carbon-nitrogen ratios were estimated from at least two analytical replicates of subsamples of the bulked cores from each planting site (n = 6/experimental unit)

LITERATURE CITED

Olsen, S. R., C. U. Cole, F. S. Watanabe, and L. A. Dean. 1954. Estimation of available phosphorus in soils by extraction with sodium bicarbonate. USDA Circular 939:19.



[Back to E086-129]