Premium
Towards an extended framework for the general dynamic theory of biogeography
Author(s) -
Carey Mark,
Boland John,
Weigelt Patrick,
Keppel Gunnar
Publication year - 2020
Publication title -
journal of biogeography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.7
H-Index - 158
eISSN - 1365-2699
pISSN - 0305-0270
DOI - 10.1111/jbi.13944
Subject(s) - archipelago , species richness , biogeography , ecology , taxon , insular biogeography , biodiversity , akaike information criterion , geography , global biodiversity , cape verde , species diversity , biology , statistics , mathematics , history , ethnology
Aim The General Dynamic Model of island biogeography (GDM) explains patterns of species richness on volcanic hotspot archipelagos with respect to island age ( T ) and area ( A ), most commonly employing what we here call semi‐log ATT 2 and LogATT 2 models. These can be considered direct extensions of semi‐log and power law species‐area relationships (SARs), with a hump‐shaped function of T modelling drivers of biodiversity over and above area per se. We compare these and related models and investigate their conceptual and mathematical connections. Location The Azores, Galapagos, Hawaii, Marquesas, Society Islands, Cape Verde Islands, Canary Islands and Aeolian archipelagos. Taxa Vascular plants, birds, bryophytes, arthropods and molluscs. Methods We gathered species richness data from databases and published sources, resulting in 34 datasets for various taxa. We used the Akaike information criterion corrected for small sample size (AICc) to compare the relative performance of models. We analysed the data at archipelago level and across archipelagos by using mixed effect models. We also calculated the relative time of expected peak diversity ( T peak ) from parameter estimates. Results In almost three‐fourth of the cases the best model overall was one that contained only A or T rather than both predictors. There was greater variance between taxa than between archipelagos. Better dispersed taxa tended to reach peak diversity earlier. The most generally applicable of the models including T were the LogATT 2 and LogTT 2 models. These have important, previously unrecognized mathematical properties. Main Conclusions SARs and species–time relationships should always be considered as potentially more parsimonious options. Considering the LogATT 2 model as a modification of the power law S = cA z , where a lognormal function in T replaces the constant, c allows the construction of an extended framework for the GDM with incorporation of additional predictors such as measures of environmental heterogeneity and application beyond hotspot archipelagos.