Year 2026, Volume 72 Issue 2
| Year : | 2026 |
| Volume : | 72 |
| Issue : | 2 |
| Authors : | Karla Mayara ALMADA GOMES; Antonio Pedro Fragoso WOYCIKIEVICZ; Diego dos Santos Vieira CARVALHO; Carla Talita PERTILLE; Lizandra Elizeário dos SANTOS; Renato Bezerra da Silva RIBEIRO; Thiago de Paula PROTÁSIO; Afonso FIGUEIREDO FILHO |
| Title : | MODELING LATEX AND CERNAMBI PRODUCTION IN AMAZONIAN Hevea guianensis Aubl. PLANTATIONS |
| Abstract : | The sustainable management of native and planted rubber tree stands is essential to optimize natural rubber production and conserve this forest resource. This study assessed the productive performance of rubber tree plantations in the Tapajós National Forest and developed statistical models to estimate latex and cernambi (coagulated natural rubber) production, considering the ecological and structural characteristics of Amazonian stands. A full inventory of all trees was conducted in both areas, recording diameter at breast height (DBH), total height and commercial height. Latex and cernambi production were measured weekly over one year in sampled trees distributed across all observed DBH classes. Additional dendrometric and morphometric variables were also measured in these trees. Correlations were analyzed between dendrometric variables, morphometric indices and combined variables. Statistical modelling included: (1) Spearman correlation, (2) stepwise selection using Variance Inflation Factor and (3) stepwise selection based on Akaike Information Criterion. High structural variability was observed. Trees with DBH≥60 cm showed the highest latex production, while those with DBH≥40 cm showed higher cernambi production. DBH and the composite variable D²h were the strongest predictors, with significant positive correlations. The selected models showed moderate explanatory capacity, with R²adj values of 0.309 for latex and 0.291 for cernambi, and relative standard errors of 26.24% and 23.69%, respectively, which is consistent with the high structural and environmental heterogeneity characteristic of Amazonian forest-based rubber systems. Cross-validation results confirmed stable predictive performance, with R²cv values of 0.294 and 0.259 for latex and cernambi, respectively. The proposed models provide quantitative support for estimating latex and cernambi production in Amazonian rubber plantations, with potential application in community-based management systems for non-timber product extraction. |
| For citation : | Almada Gomes, K.M., Woycikievicz, A.P.F., Carvalho, D.S.V., Pertille, C.T., Santos, L.E., Ribeiro, R.B.S., Protásio, T.P., Figueiredo Filho, A. (2026). Modeling latex and cernambi production in Amazonian Hevea guianensis Aubl. plantations. Agriculture and Forestry, 72 (2): 351-378. https://doi:10.17707/AgricultForest.72.2.22 |
| Keywords : | non-timber forest products; Amazonian rubber plantations; statistical modelling; latex production; cernambi production; Hevea guianensis |
| download paper | |