An empirical function for predicting survival over a wide range of densities

Abstract

A flexible sigmoid function of the form S = 1/[1 + (H/Hc)⁰], where S is survival expressed as the ratio of number of living trees (N) to the number of trees at stand  establishment (Ni), H is mean stand height, H_c is mean height at S = 0.5, and 0 is an exponent defining the shape of the curve, was fit to a set of loblolly pine survival data cover­ing a range of densities from 1000 to 16,000 trees per acre. The equation provided an adequate fit to the data vith r² values from 0.76 to 0.98. The estimated parameters H_c and 0 were found to be correlated to initial density. Both can be predicted satisfactorily with regressions of the form H_c= b₀N_i ^b1, and 0 = b₀ + b_i/N_i.

Citation

Harms, Williams R. 1983. An empirical function for predicting survival over a wide range of densities. Charleston, SC. In: Proceedings of the Second Biennial Southern Silvicultural Research Conference; 1982 November 4-5; Atlanta, GA: p.334-337.