Testing the hypothesis of hierarchical predictability in ecological restoration and succession

摘要: To advance predictive ecology, the hypothesis of hierarchical predictability proposes that community measures for which species are interchangeable (e.g., structure and richness) more predictable than identity matters composition). Predictability is hypothesized to decrease response in order following categories: structure, richness, function, composition. We tested this using a 14-year, oak savanna-prairie restoration experiment removed non-native pine plantations at 24 sites northwestern Ohio, USA. Based on measures, data showed minimal support hypothesis, because varied within categories. Half had over half their variability modeled fixed (restoration treatment year) random plot effects, these "predictable" occurred all four Pine basal area, environment soil texture), antecedent vegetation accounted variation change first three post-restoration years 77% measures. Change between 3rd 14th was less predictable, but most increased favorably via achieving them unique ways. propose will not conform with ecosystems dynamics driven by stochastic processes such as seed dispersal, or where richness influenced The ability predict measure may be number combinations casual factors affecting values it can have.