5. STUDY METHODOLOGY AND ANALYSIS OF DATA

In order to investigate the long-term impacts of harvesting operations on the environment, data were obtained by remeasurement of Plots B/F09 (“conventional” logging) and B/G09 (“environmentally sound” logging), which had been harvested four years prior to the remeasurement. This involved the following steps:

Qualitative and quantitative assessments were made of the regeneration of commercial and non-commercial tree species on skidtrails and in gap openings in the two harvested plots.

A qualitative assessment of soil compaction was made on primary and secondary skidtrails as well as on undisturbed soil. Primary skidtrails are skidtrails that were used multiple times for the skidding of several trees, whereas secondary skidtrails are those that served only for the skidding of one or two trees that were accessed individually. Secondary skidtrails are found only in Plot B/F09, as the environmentally sound harvesting system used by PWA precludes the use of skidtrails to individual trees.

An assessment was made of the status of PCTs, including an estimation of the present wood volume in the two harvested plots. Similarly, all residual trees of commercial species were assessed, both for their current volume and for damage received during the logging operation four years previously.

In addition to these environmental assessments, the structural development in the forestry sector and the processing sector at PWA were analysed as well as the general development of the PWA project over the years following its pre-operational test phase, which ended in 1997.

5.1 Assessment of regeneration

Post-harvest assessments of the biotic impact of harvesting operations and the regeneration of harvested areas provide basic information on the future economic and ecological potential of the forests and the sustainability of their use. The composition of tree species in the forest plots gives an impression of the future stability and potential economic value of the forests.

The regeneration of both commercial and non-commercial tree species was examined as well as the regeneration of competing vegetation such as climbers and palms. As the sample plots had been harvested four years prior to the assessment, it can be assumed that sufficient time had passed to provide a representative image of the regeneration even of non-pioneer tree species.

The species of every tree seedling, sapling, and pole within the sampling plots was recorded along with its diameter at a height of 10 cm from the ground. In addition, data were recorded on the presence of competing vegetation such as climbers and palms.

5.1.1 Assessment of regeneration in canopy gaps

Regeneration in gap openings was investigated in both cutting units B/F09 and B/G09. The assessments in each gap opening were conducted in five sampling plots, each measuring 2 m × 2 m. Altogether, 40 regeneration plots were assessed in eight gap openings, four in each of the two cutting units.

The location of individual regeneration plots depended on the form of the gap opening. In “regularly shaped” gaps formed by the removal of a single tree in which an apparent felling direction could be determined, the regeneration plots were distributed as Winkler (1997) proposed, along the apparent felling direction of the tree at a spacing of 10 m between adjacent regeneration plots. After four years of regeneration, however, the majority of the gaps were shaped rather irregularly and it was not possible to determine an apparent felling direction. In such gaps the five regeneration plots were distributed evenly throughout the area of the gap.

The results of the assessments are summarised for the two forest plots in Tables 8 and 9. Because of the high degree of variability in the regeneration data, no statistically valid correlation between gap size and regeneration frequency could be found. An apparent positive correlation between gap size and the concentration of palms in the sampling plots of cutting unit B/G09 was also not statistically significant.

Overall, the regeneration densities of both commercial and non-commercial tree species are quite similar in the two cutting units. The average number of climbers/m² was higher in cutting unit B/F09, whereas a higher number of palms/m² was found in cutting unit B/G09.

Table 8. Density of regeneration in canopy gaps of Plot B/G09 (“environmentally sound” logging treatment).

Table 9. Density of regeneration in canopy gaps of Plot B/F09 (“conventional” logging treatment).

Tables 10 and 11 provide a detailed summary of the number of different species of seedlings, saplings, and poles of commercial and non-commercial trees that were found in the gap openings of the two cutting units. In the four gaps investigated in cutting unit B/G09, an average of 8.25 different commercial tree species were found. The variety of tree species was slightly higher in cutting unit B/F09 with an average of 10.75 commercial species per gap opening. The average diameters of the plants were quite similar in the two cutting units, with 1.05 cm in cutting unit G/G09 and 1.08 cm in cutting unit B/F09. Because of variability in the data, differences between the two plots in the number of species per gap opening and in the average diameter of measured tree regeneration are not statistically significant.

Table 10. Composition of tree regeneration (seedlings, saplings, and poles) in canopy gaps assessed in Plot B/G09 (“environmentally sound” logging treatment). Diameters were measured 10 cm above the ground.

Table 11. Composition of tree regeneration (seedlings, saplings, and poles) in canopy gaps assessed in Plot B/F09 (“conventional logging” treatment). Diameters were measured 10 cm above the ground.

5.1.2 Assessment of regeneration on skidtrails

As the skidtrails in the PWA project are maintained permanently with the intention that they should be reused during subsequent harvesting cycles, an assessment of regeneration on skidtrails is of little relevance to the future properties of the forests in the PWA project. Still, the assessment is reported here to give an impression of the influence on regeneration of heavy soil compaction by harvesting machines in the yellow latosols that are common in this part of the Amazon Basin. The influence of soil compaction on regeneration is of major importance in forests where the skidtrails are not maintained as part of the permanent infrastructure. This is the common practice throughout the Amazon region.

Regeneration was investigated on skidtrails in both cutting units. In unit B/G09, assessments were made in 20 sample plots on the two existing primary skidtrails. In unit B/F09, assessments were performed on 10 sampling plots on a primary skidtrail and on 10 sampling plots on a secondary skidtrail.

The regeneration inventories were made with 2 m × 2 m sample plots placed across each skidtrail at intervals of 30 m along the skidtrail. Each inventory plot crosses three different soil conditions: side strips, ruts, and the middle part of the skidtrail. Data were therefore recorded separately for these different soil conditions, as proposed by Winkler (1997).

Following the methodology used in the gap openings, the species of every seedling, sapling, and pole with a height of at least 10 cm was recorded and its diameter was measured at a point 10 cm above the ground. The frequencies of occurrence of climbers and palms were recorded as well.

In both cutting units, the regeneration of all types of vegetation, commercial and non-commercial tree species as well as competitive plants, show clear differences in the parts of the skidtrails representing different degrees of soil compaction. The density of regeneration strongly decreases from the less compacted side and middle strips to the highly compacted ruts.

Table 12. Density of regeneration on primary skidtrails in Plot B/G09 (“environmentally sound” logging treatment).

Results for the primary and secondary skidtrails assessed in cutting unit B/F09 can be found in Tables 13 and 14. As with the results obtained for cutting unit B/G09, the average density of vegetation was significantly lower in the ruts than in the side and middle strips of the skidtrails. The higher density of regeneration expected on secondary skidtrails due to potentially lower soil compaction was not found. This suggests that even a single trip over a skidtrail causes significant soil compaction.

Table 13. Density of regeneration on a primary skidtrail in Plot B/F09 (“conventional logging” treatment).

Table 14. Density of regeneration on a secondary skidtrail in Plot B/F09 (“conventional logging” treatment).

In order to estimate the severity of soil compaction on skidtrails, infiltration rates were measured by means of an infiltrometer produced locally with simple pipes according to the procedure recommended by Hofmann (1989). The measurements were performed after a period of 10 days of tropical rain, so the soils were already saturated and infiltration rates were low. Following the methodology used for the assessment of regeneration on skidtrails, the measurements were performed on skidtrails in both cutting units. The positions of the regeneration inventories were maintained and the infiltration rates were determined separately on side strips, the ruts, and the middle part of every strip. The resulting data are summarised in Table 15.

Table 15. Water infiltration rates measured on skidtrails, in mm/h. Plot B/G09 was harvested with “environmentally sound” logging and B/F09 with “conventional” logging.

5.2 Assessment of PCTs and other harvestable trees

The general health status and damage to stems and crowns of residual trees of commercial species four years after harvesting operations can be taken as an indicator of the future development of the forest. For this purpose, data were collected as follows:

1. For PCTs, i.e., trees of commercial species with DBH from 20 cm to 50 cm, detailed assessments were made in one subplot of 2.25 ha (150 m × 150 m) within each cutting unit B/F09 and B/G09.

2. All residual trees of commercial species with DBH ≥ 50 cm were assessed over the entire area of 10 ha in each cutting unit B/F09 and B/G09.

Table 16 summarises results of the assessment of PCTs for the subplots within the two cutting units. By definition PCTs include only healthy trees that are expected to form the harvest crop during the next harvest entry, so damaged trees that otherwise met the definition of PCTs were excluded. There were no significant differences between the results for the two subplots.

Table 16. Characteristics of PCTs (DBH 20–50 cm). Subplot B/F09 was harvested with “conventional” logging and Subplot B/G09 with “environmentally sound” logging.

Residual trees of commercial species with DBH ≥ 50 cm were assigned to one of three damage classes according to the condition of their stems and crowns as follows:

Data for residual trees of commercial species with DBH ≥ 50 cm are shown in Table 17. The total number of residual trees shows a great difference in the two cutting units—the number is twice as high in Plot B/G09, treated with “environmentally sound harvesting,” as in Plot B/F09, treated with “conventional logging”. The comparisons of residual wood volume show even larger differences, as the mean diameter and thus the average volume per tree is higher in Plot B/G09 as well.

Similarly, damage to stem and crown was significantly worse in Plot B/F09 than in Plot B/G09. The difference between the plots is especially large for trees in damage class 3, representing severe damage to the crown and stem.

Table 17. Characteristics of residual trees of commercial species with DBH ≥ 50 cm. Plot B/F09 was harvested with “conventional” logging and Plot B/G09 with “environmentally sound” harvesting.