Human-induced global heating is significantly impacting global crop yields worldwide, driven by mechanisms such as drought, waterlogging, extreme-weather events, and new crop pest patterns. This threatens the UN's sustainable development goal to end hunger by 2030 (SDG 2: Zero Hunger). To mitigate negative impacts on food production and to ensure food security, agricultural, climate and biodiversity research must develop and implement integrative adaptive strategies. For this, it is crucial to integrate and use data from diverse sources, such as high-resolution earth observation (EO), biodiversity monitoring, gene banks and socio-economic data. Increasing usage of underutilized crops and crop wild relatives (CWR) offers a promising approach for food security due to their greater genetic diversity and resilience to more variable environments than conventional crops. In the context of BioDT (https://biodt.eu/use-cases/crop-wild-relatives-and-genetic-resources-food-security), we are developing a digital twin for crop genetic resource modeling. This digital twin combines known species occurrences with actual climate data from EO for advanced species distribution modeling (SDM). The aim is to generate habitat suitability maps and identify potential areas where relevant populations of CWRs with interesting genotypes can be found. We deploy our SDM model within the Destination Earth Platform (DEDL Islet service) to leverage its near-data processing capabilities and to incorporate internal and federated EO data, such as ERA5-Land/CDS and the Climate Change Adaptation Digital Twin to enhance the precision of our model. Therefore, we developed a platform around the Argo workflow engine also used by the DestinE Hook service, enabling seamless integration of our workflows within the DestinE ecosystem. Building on the FAIR (Findable, Accessible, Interoperable, Reusable) data principles, our platform demonstrates an end-to-end integration of FAIR practices within the Destination Earth platform. We utilize RO-Crates, a flavor of FAIR Digital Objects (FDOs) and established vocabularies from schema.org/Bioschemas to describe our data and provide provenance, thereby ensuring interoperability and reproducibility across digital data spaces. Finally, by leveraging FAIR signposting, we make these FDOs "webby", ensuring they are discoverable and machine-actionable by software agents.