To model distribution ranges, the most popular methods of phylogenetic biogeography divide Earth into a handful of predefined areas. Other methods use explicit geographic ranges, but unfortunately, these methods assume a static Earth, ignoring the effects of plate tectonics and the changes in the landscape. To address this limitation, I propose a method that uses explicit geographic ranges and incorporates a plate motion model and a paleolandscape model directly derived from the models used by geologists in their tectonic and paleogeographic reconstructions. The underlying geographic model is a high-resolution pixelation of a spherical Earth. Biogeographic inference is based on diffusion, approximates the effects of the landscape, uses a time-stratified model to take into account the geographic changes, and directly integrates over all probable histories. By using a simplified stochastic mapping algorithm, it is possible to infer the ancestral locations as well as the distance traveled by the ancestral lineages. For illustration, I applied the method to an empirical phylogeny of the Sapindaceae plants. This example shows that methods based on explicit geographic data, coupled with high-resolution paleogeographic models, can provide detailed reconstructions of the ancestral areas but also include inferences about the probable dispersal paths and diffusion speed across the taxon history. The method is implemented in the program PhyGeo.

中文翻译：

---

使用动态古地理模型和明确的地理范围进行系统发育生物地理学推断


为了模拟分布范围，最流行的系统发育生物地理学方法将地球划分为几个预定义的区域。其他方法使用明确的地理范围，但不幸的是，这些方法假设地球是静态的，忽略了板块构造和地貌变化的影响。为了解决这个限制，我提出了一种方法，该方法使用明确的地理范围，并结合直接源自地质学家在构造和古地理重建中使用的模型的板块运动模型和古景观模型。底层地理模型是球形地球的高分辨率像素化。生物地理推断基于扩散，近似景观的影响，使用时间分层模型考虑地理变化，并直接整合所有可能的历史。通过使用简化的随机映射算法，可以推断祖先位置以及祖先谱系走过的距离。为了便于说明，我将该方法应用于无患子科植物的经验系统发育。这个例子表明，基于明确地理数据的方法与高分辨率古地理模型相结合，可以提供祖先地区的详细重建，但也可以包括对整个分类单元历史上可能的扩散路径和扩散速度的推断。该方法在 PhyGeo 程序中实现。