A survival model for individual shortleaf pine trees in even-aged natural stands
A model was developed that predicts the probability of survival for individual shortleaf pine (Pinus echinata Mill.) trees growing in even-aged natural stands. Data for model development were obtained from the first two measurements of permanently established plots located in naturally occurring shortleaf pine forests on the Ouachita and Ozark National Forests in western Arkansas and eastern Oklahoma. The logistic function was used to model survival probability. Parameters were estimated by using logistic regression in which the dependent variable was "1" for trees alive during both inventories and "0" for trees that died prior to the second inventory. Examination of several combinations of independent variables (representing tree size, relative position of the tree in the stand, stand density, and stand age) yielded the following model:
POS = (1 + exp(-(b0 + b1/DR + b2SBA + b3DH)))-1
where POS is annual probability of survival, DR is the ratio of quadratic mean diameter to individual tree d.b.h., SBA is stand basal area in square feet per acre, DH is average height of dominant and codominant trees in feet, b0, b1, b2, b3 are parameter estimates. A chi-square evaluation was performed to test model performance. This tree survival model is being used to estimate probability of individual tree survival in a distance-independent individual tree simulator for shortleaf pine.