|Lange, H; Thies, B; Hauhs, M; Kastner-Maresch, A: Recurrence quantification of transient behavior in simulated evolving forests, Proceedings of the German Workshop on Artificial Life, 83-96 (1998)|
We analyze the behavior of observable quantities of the detailed individual-oriented tree growth model TRAGIC++. The model is run on very large time scales (50 kiloyears (kys)) to investigate possible evolutionary processes related to height growth strategies and on small spatial scales (support for approx. 30 adult trees). Transient phases of maximal height and tree number are investigated by means of recurrence quantification analysis (RQA). Simulations are compared with model runs where evolution is excluded. The causes behind phenotype changes in terms of strategy parameters of trees can be identified with the help of RQA, while linear methods alone give ambiguous results. Our investigations suggest that RQA might be a suitable tool for detecting causal relationships in noisy and complex environments.