A plea for open source greenhouse climate modelling

Process based modelling of the greenhouse climate and crop has been used for over 40 years to study, design, and control greenhouse cropping systems. In recent years, a multitude of new models has been developed, but many of these newly developed models are essentially copies of legacy models that were developed in the 1990s. A question that arises is why this is the case. We argue that a lack of transparency and a culture of closed science practices leads to a repetitive habit of reinventing the wheel. An analysis of published models, both from recent years and from previous decades, shows that model reuse is often confined to the limits of isolated research groups. This is especially true for large models that incorporate new and complex climate control equipment and energy saving techniques. At the same time, the greenhouse sector is undergoing rapid development and facing momentous challenges. Process-based modelling can provide key insights and help address these challenges, but essential resources are being lost to re-coding and re-programming old models. We suggest that a collaborative research culture in greenhouse modelling – one that fosters free software, open source code, open data and shared benchmarks – is essential for the field to keep up with new developments and address current and future challenges.