A model for fire-induced heating in tree stems is linked to a recently reported model for tissue necrosis. The combined model produces cambial tissue necrosis predictions in a tree stem as a function of heating rate, heating time, tree species, and stem diameter. Model accuracy is evaluated by comparison with experimental measurements in two hardwood and two softwood species: red maple (Acer rubrum), chestnut oak (Quercus prinus), ponderosa pine (Pinus ponderosa), and Douglas-fir (Pseudotsuga menziesii). Results are promising, and indicate that the model predicts stem mortality/survival correctly in ~75-80% of the test cases. A limited sensitivity analysis of model kill depth predictions suggests that the model is more sensitive to required input data for some species than for others, and that the certainty in predicting vascular cambium necrosis decreases as stem diameter decreases.