<i>N‐SDM</i> : a high‐performance computing pipeline for Nested Species Distribution Modelling

Predicting contemporary and future species distributions is relevant for science decision making, yet the development of high-resolution spatial predictions numerous taxonomic groups regions limited by scalability available modelling tools. Uniting distribution (SDM) techniques into one high-performance computing (HPC) pipeline, we developed N-SDM, an SDM platform aimed at delivering reproducible outputs standard biodiversity assessments. N-SDM was built around a spatially-nested framework, intended facilitating combined use occurrence data retrieved from multiple sources various scales. allows combining two models fitted with covariate global to regional scales, which useful addressing issue niche truncation. The set state-of-the-art features embodied in includes newly devised selection procedure, five algorithms, algorithm-specific hyperparameter grid search, ensemble small-models approach. designed be run on HPC environments, allowing parallel processing thousands same time. All information required installing running openly GitHub repository https://github.com/N-SDM/N-SDM.