Fig 1.
Distribution of Brazil nut concessions in Madre de Dios, Peru, with approximate locations of the 135,528 individual Brazil nut trees (DBH≥10 cm) from 418 concessions considered in this paper.
Fig 2.
Relationship between average seed production of individual Brazil nut trees measured over two years by Rockwell et al. [9] and seed production estimates by Brazil nut harvesters.
Only trees that could be matched with a precision of ~10m (n = 86), ~20m (n = 121) and ~40m (n = 166), respectively, and had a similar diameter (10% error margin) were considered out of a total of N = 457 trees with unique coordinates measured by Rockwell et al. [9]. The regression lines are in red and 95% confidence intervals are indicated by grey polygons.
Fig 3.
a. Cumulative contribution to overall estimated seed production of all 135,528 trees in our database ordered by their individual production estimates b. Diameter distribution of trees that were said by Brazil nut harvesters to never or hardly ever produce seeds.
Table 1.
Relations between seed production of Brazil nut trees and a number of phenotypic, phytosanitary and environmental variables described in literature and/or reported by Brazil nut concessionaires and harvesters, compared with those obtained from seed production estimates sourced from harvesters in Madre de Dios.
Fig 4.
Relations between estimated seed production of Brazil nut trees and (a) stem height to the first branch (crown position); (b) crown height; (c) total tree height; (d) DBH; (e) above ground woody biomass based on DBH and height measurements; and (f) above ground woody biomass based on DBH only. Red lines represent GAMM smoothers. Dashed lines show 95% confidence intervals.
Fig 5.
Boxplots comparing Brazil nut seed production estimates between trees with and without (a) broken branches; (b) evidence of wound exudate; (c) holes in their trunks; (d) presence of termite nests; (e) presence of tumours; (f) liana infestation. Statistical comparisons are based on Wilcoxon tests (n = 55,644).
Fig 6.
Relation between estimated seed production of Brazil nut trees and the distance to the three nearest conspecific trees.
The solid red lines represent GAM smoothers. Dashed lines show 95% confidence intervals.