Parameter calibration of climate suitability models is often hindered by the lack of absence data for plant species, limiting their effectiveness for global scale applications. Here we propose a novel calibration approach based on the Generalized Likelihood Uncertainty Estimation (GLUE) framework that eliminates the needs for background samples. This method defines the likelihood statistics under the assumption that the distribution of the climate suitability index at the occurrence sites differs from that across all locations within a given region. To prioritize presence data, a weighted likelihood function was incorporated into the GLUE procedure. We demonstrated the utility of this approach through a case study on orange (Citrus sinensis), a crop whose climate suitability has rarely been evaluated at a global scale. Model performance improved when the parameter search spaces were defined with minimal ecological constraints, which resulted in a clear separation between producing and non-producing countries. These findings suggest that the proposed approach offers a robust and scalable alternative for climate suitability modeling in data-sparse contexts. This framework is broadly applicable to both cultivated and invasive species, enabling reliable projection of the potential distribution to inform land-use planning and climate adaptation strategies.
Hyun et al. (2026) studied this question.