Comprobación de cinco métodos de muestreo forestal en un bosque nativo de Araucaria angustifolia Bert. O. Ktze

in a native forest of araucaria angustifolia belonging to the national forest of san francisco of paula (rs-br), was carried out a comparison of four methods of sampling of variable area (strand, prodan, quadrants and bitterlich) with a method of fixed area in order to determine the efficiency of each one to estimate the quantitative parameters (volume, number of trees by hectare, basal area) and the qualitative (number of species) of the population. the inventory was carried in one hectare considering all the trees with circumference to the height of the chest (cap) 3 30 cm, identifying them, and measuring their total and commercial heights. the adopted method of fixed area was that of the strips of 100 m long for 10 m wide (10 contiguous strips). the prodan, bitterlich and quadrants methods were carried out on 25 sampling points distributed 20 m systematically each, and for the method of strand were 30 lines of sampling of 15,7 m long selected. for analysis of the variance it was proved that the methods did not present significant differences in the estimate of the volume, basal area and number of trees by hectare. there were significant differences in the number of species, and the fixed area method presented larger stocking. the relative efficiency was calculated based in the timing and the variation coefficients thus the strand method was the best to estimate all the parameters, whereas the quadrants method under estimate all of them.