Ecological Archives A019-058-A1

Lilian Blanc, Marion Echard, Bruno Herault, Damien Bonal, Eric Marcon, Jérôme Chave, and Christopher Baraloto. 2009. Dynamics of aboveground carbon stocks in a selectively logged tropical forest. Ecological Applications 19:1397–1404.

Appendix A. Detailed site description.

Description of Site

All inventories were conducted at the Paracou experimental site (5o 18' N, 52o 55 'W), a lowland tropical rain forest near Sinnamary, French Guiana (Gourlet-Fleury et al. 2004). The forest is typical of Guianan rainforests (ter Steege et al. 2003). More than 550 woody species attaining 2 cm DBH have been described at the site, with an estimated 160–180 species of trees ≥ 10 cm dbh per hectare (Molino and Sabatier 2001; Baraloto et al., unpublished data). The dominant families at the site include Leguminoseae, Chrysobalanaceae, Lecythidaceae, Sapotaceae, and Burseraceae.

In 2007, stem density of trees larger than 10 cm in diameter at breast height (DBH) varies from 564 – 670 ha-1 with a mean of 609 ha-1. Basal area varies from 28.1 – 31.8 m2 ha-1 with a mean of 30.4 m2 ha-1.

Description of Plots and Inventories

In 1984, 48 square 1.56 ha plots delimited by 1m wide were established at the Paracou site for a complete inventory of all trees ≥ 10 cm DBH. Trees were originally identified into 58 species groups that represented classes of commercial species, as the initial goal of the project was to determine rotation cycles for timber species (Gourlet-Fleury et al. 2004). For species that were harvested, these groups are precise to the taxonomic level of genus in most cases. In most other cases, family-level groupings were made (e.g., Chrysobalanaceae, Lecythidaceae, Arecaceae). Since 1992, this list has been expanded to more than 200 groups, and as of 2006, 74 to 94% of living stems per parcel have been assigned to at least the taxonomic level of genus.

Annual inventories of mortality, recruitment and diameter growth were conducted until 1995 and every two years thereafter. Here we report data until the most recent inventory in 2007 (Fig. A1). Diameter (DBH) was calculated from circumference measures made to a precision of 0.5 cm. For trunks without buttresses, measurements were made at positions of 1.3 m above the soil surface and points of measurement were marked using permanent paint. For trunks with buttresses, these measures were moved to a position 10 cm above the buttresses and also marked with paint. All trees were kept in data analysis for both biomass and growth estimations except those with a negative diameter growth ≤ -1.5 cm.yr-1. The mean biomass of this subset of trees accounted for 2.7% (min: 1.2; max: 4.2) of the total biomass in each parcel.

Description of Logging Activities and Silvicultural Treatments

From 1986 to 1988, the plots underwent three logging treatments according to a randomized block design, with three replicate blocks of twelve plots, assigned as controls or one of three treatments. A precise description of the Paracou site is given by Gourlet-Fleury et al. (2004), but the following is a brief summary of the treatments. In Treatment 1, selected timbers were extracted, with an average of 10 trees ≥ 50 or 60 cm dbh removed per hectare. Treatment 2 was logged as in Treatment 1, followed by timber stand improvement (TSI) by poison girdling of selected non-commercial species, with about 30 trees ≥ 40 cm dbh removed per hectare. Treatment 3 was logged as in Treatment 2 for an expanded list of commercial species, with about 45 trees ≥ 40 cm dbh removed per hectare. Depending on stand composition, some of the latter plots were less intensively logged than treatment 2 plots. We thus have grouped treatments 2 and 3 in our analyses as plots receiving TSI, within which we examine the direct effect of logging intensity with estimates of timber volume removed.

Tree harvesting in the plots was initiated in October 1986 and was completed in May 1987. Timber stand improvement by poison girdling began in December 1987.

Characterization of logging intensity

Table A1 presents the broad range of variability in four variables describing logging intensity at Paracou. We calculated the amount of wood harvested in two ways. First, we used a count of the number of trees that were cut and for which the logs were skidded and removed from the forest. Second, we calculated the volume of wood that was removed from the forest. Every 2m section of roundwood harvested at Paracou was measured for basal and apical circumference (Fig. A2) and its fate was recorded as left in the forest or skidded to the log landing with subsequent transport to the sawmills. These data were used to calculate the volume of wood removed from each forest plot, as the sum of all pieces that were skidded.

We also used a geographic information systems (GIS) spatial database that was constructed using detailed maps drawn at the time of harvesting activities. In particular, two types of information were mapped that also measure harvest intensity. First, all skid trails were mapped and converted to a shapefile in ArcGIS 9 (ESRI, Redlands, CA). Primary and secondary skid trails were not distinguished in these maps. Skid trail area in the logged 1.56 ha plots ranged from 811 to 3616 m2. In addition, areas of canopy opening due to tree felling were also projected to the ground surface and converted to an ArcGIS shapefile. Harvest gap area in the logged 1.56 ha plots ranged from 2034 – 8977 m2.

The correlation between the number and volume of harvested trees indicates the consistency of the size of trees that were harvested (Pearson correlation coefficient: 0.74 and 0.80 for plots without and with TSI, respectively). Given the high degree of correlation among these two variables, we chose to present data on the effects on carbon flux only for the variable that is most often described for logging intensity, i.e., the Volume Skidded.

Overall, the Paracou plots represent a broad gradient of harvest intensities. In addition, they can be characterized as having higher damage rates per harvest intensity than other sites conventionally harvested (CL), and much higher rates than sites harvested at similar intensities using reduced impact logging (RIL) techniques (Fig. A3).

 

TABLE A1. A summary of the four variables used to characterize the 36 logged plots (1.56 ha each; n = 24 for plots receiving timber stand improvement, TSI). Values are reported on a per-hectare basis.

 

Trees cut

Volume skidded (m3)

Proportion of area in skid trails

Proportion of area in logging gaps

Basal area poisoned
(m2)

 

Log

Log + TSI

Log

Log + TSI

Log

Log + TSI

Log

Log + TSI

Log

Log + TSI

mean

10.4

20.6

32.51

53.42

0.12

0.14

0.22

0.30

-

4.6

min

5.8

5.1

15.38

12.43

0.07

0.05

0.13

0.14

-

1.11

max

15.4

41.6

51.85

109.83

0.23

0.23

0.33

0.58

-

8.30

CV

30%

56%

38%

51%

38%

39%

30%

33%

-

37%

 

FigA1
 
   FIG. A1. Changes in stand-level average diameter increment and annualized mortality and recruitment rates, in the Paracou plots. Logging took place in 1986 and ended in 1987.

 

FigA2
 
   FIG. A2. Illustration of the fractions of aboveground biomass including those calculated for harvested trees.

 

FigA3
 
   FIG. A3. A qualitative characterization of the logging practices at the Paracou plots relative to other forestry operations described in the literature. In general, damage to forest stands increases with increasing logging intensity (e.g., timber volume extracted). Under conventional logging techniques, this damage is much higher than under reduced-impact logging. The Paracou plots are representative of the most damaging conventional logging practices across a broad gradient of harvest intensities (see Table A1). Adding timber stand improvement (TSI) treatments appears to increase damage per unit of timber volume harvested, but it remains unclear how applying TSI to sites with RIL will affect forest damage and carbon dynamics.

 

LITERATURE CITED

Gourlet-Fleury, S., J.-M. Guehl, and O. Laroussinie, editors. 2004. Ecology and management of a neotropical forest. Lessons drawn from Paracou, a long-term experimental resarch site in French Guiana. Elsevier, Paris, France.

Molino, J. F., and D. Sabatier. 2001. Tree diversity in tropical rain forests: A validation of the intermediate disturbance hypothesis. Science 294:1702–1704.

ter Steege, H., N. Pitman, D. Sabatier, H. Castellanos, P. Van der Hout, D. C. Daly, M. Silveira, O. Phillips, R. Vasquez, T. Van Andel, J. Duivenvoorden, A. A. De Oliveira, R. Ek, R. Lilwah, R. Thomas, J. Van Essen, C. Baider, P. Maas, S. Mori, J. Terborgh, P. N. Vargas, H. Mogollon, and W. Morawetz. 2003. A spatial model of tree alpha-diversity and tree density for the Amazon. Biodiversity and Conservation 12:2255–2277.



[Back to A019-058]