In recent decades, finite-element modelling (FEM) has become a very popular tool in volcanological studies and has even been used to describe complex system geometries by accounting for multiple reservoirs, topography, and heterogeneous distribution of host rock mechanical properties. In spite of this, the influence of geological information on numerical simulations is still poorly considered. In this work, 2D FEM of the Colima Volcanic Complex (Mexico) is provided by using the Linear Static Analysis (LISA) software in order to investigate the stress field conditions with increasingly detailed geological data. By integrating the published geophysical, volcanological, and petrological data, we modelled the stress field considering either one or two magma chambers connected to the surface via dykes or isolated (not connected) in the elastic host rocks (considered homogeneous and non-homogeneous). We also introduced tectonic disturbance, considering the effects of direct faults bordering the Colima Rift and imposing an extensional far-field stress of 5 MPa. We ran the model using the gravity in calculations. Our results suggest that an appropriate set of geological data is of pivotal importance for obtaining reliable numerical outputs, which can be considered a proxy for natural systems. Beside and beyond the importance of geological data in FEM simulations, the model runs using the complex feeding system geometry and tectonics show how the present-day Colima volcanic system can be considered in equilibrium from a stress state point of view, in agreement with the long-lasting open conduit dynamics that have lasted since 1913.

中文翻译：

---

通过结合有限元建模（FEM）模拟和地质数据分析科利马火山群（墨西哥）的应力场条件

在最近的几十年中，有限元建模（FEM）已成为火山学研究中非常流行的工具，甚至通过考虑多个储层，地形和宿主岩力学特性的非均质分布，甚至被用来描述复杂的系统几何形状。尽管如此，仍很少考虑地质信息对数值模拟的影响。在这项工作中，通过使用线性静态分析（LISA）软件提供了科利马火山群（墨西哥）的2D有限元分析，以便利用越来越详细的地质数据来研究应力场条件。通过整合已发布的地球物理，火山学和岩石学数据，我们通过一个或两个通过堤坝连接到地表的岩浆室或在弹性主岩中隔离（未连接）的岩浆腔（考虑了均质和非均质）对应力场进行了建模。考虑到与科利马裂谷接壤的直接断层的影响并施加了5 MPa的扩展远场应力，我们还引入了构造扰动。我们在计算中使用重力来运行模型。我们的结果表明，适当的一组地质数据对于获得可靠的数值输出至关重要，可以将其视为自然系统的代理。除了有限元模拟中地质数据的重要性外，