The growing impact of climate change has led to a noticeable rise in the frequency and intensity of natural hazards such as landslides, debris flows, and snow avalanches. These events are increasingly driven by extreme weather and shifting environmental conditions, posing substantial risks to critical infrastructure. A lot of these critical infrastructure are nearing the end of their intended service life, therefore designs of resilient, next-generation replacements are needed. However, current engineering guidelines may be outdated and require revision based on state of the art, physics-based numerical models that can accurately reflect evolving hazard scenarios. This work presents a numerical study of the forces generated by rapid mass movements impacting rigid and flexible protective structures, such as retaining walls and nets. The Material Point Method (MPM) is used due to its ability to model large deformations. A key challenge addressed here is the presence of oscillatory behaviour of the reaction forces upon impact, which hinder the convergence of coupled simulations. These oscillations are often non-physical, arising from numerical artifacts common in MPM, such as but not limited to, cell-crossing errors, quadrature errors, small cut instabilities and mapping errors. The influence of these artifacts on force predictions are investigated and strategies to mitigate them are explored. The findings aim to improve the reliability of structural design by enhancing the accuracy of the coupled MPM simulation of infrastructure exposed to extreme natural hazards. All the codes are implemented in Kratos Multiphysics.