Figure 1
From: An evaluation of multi-species empirical tree mortality algorithms for dynamic vegetation modelling
Comparison of model behaviour based on the projected tree cohort half-life times (MT50%, time (years) until 50% of the initially present trees died) and the response to competition (∆MT50%, i.e. the relative change of MT50% in percentage in response to a increase in BAL by 10 m2 ha−1 compared to a ‘no competition’ scenario) for six tree species at the mesic site (Bern). Note that for response to competition, positive values indicate a relative increase in mortality and negative values a decrease in mortality. Symbol size indicates the three initial tree sizes used in the simulations, and symbol color indicates MA structure (differentiating between ‘Growth-based’, ‘Competition-index’ (CI)-based and ‘Size-only’ based MAs). Numbers refer the MA sources: 1: ForClim default MA (based on theoretical assumptions9), 2: Hülsmann, et al.18 , 3:Eid and Tuhus38, 4: Monserud and Sterba39 , 5: Dursky40 , 6: Holzwarth, et al.20 , 7: Trasobares, et al.41 , 8: Crecente-Campo, et al.42 , 9: Palahi, et al.43 , 10: Bravo-Oviedo, et al.44, 11: Fridman and Ståhl45 . Results for the more xeric study site (Basel) were very similar (see Appendix S2.1).
