Agricultural digitalization is providing growing amounts of real-time digital data. Biophysical simulation models can help interpret these data. However, these models are subject to complex uncertainties, which has prompted calls for interdisciplinary research to better understand and communicate modelling uncertainties and their impact on decision-making. This article develops two corresponding insights from an interdisciplinary project in a New Zealand agricultural research organization. First, we expand on a recent Royal Society Open Science journal article (van der Bles et al. 2019 Royal Society Open Science6, 181870 (doi:10.1098/rsos.181870)) and suggest a threefold conceptual framework to describe direct, indirect and contextual uncertainties associated with biophysical models. Second, we reflect on the process of developing this framework to highlight challenges to successful collaboration and the importance of a deeper engagement with interdisciplinarity. This includes resolving often unequal disciplinary standings and the need for early collaborative problem framing. We propose that both insights are complementary and informative to researchers and practitioners in the field of modelling uncertainty as well as to those interested in interdisciplinary environmental research generally. The article concludes by outlining limitations of interdisciplinary research and a shift towards transdisciplinarity that also includes non-scientists. Such a shift is crucial to holistically address uncertainties associated with biophysical modelling and to realize the full potential of agricultural digitalization.

农业数字化正在提供越来越多的实时数字数据。生物物理模拟模型可以帮助解释这些数据。但是，这些模型受制于复杂的不确定性，这促使人们呼吁进行跨学科研究，以更好地理解和传达建模不确定性及其对决策的影响。本文从一个新西兰农业研究组织的跨学科项目中得出了两个相应的见解。首先，我们在最近扩大英国皇家学会开放科学期刊文章（范德统计局等人。2019英国皇家学会开放科学6（181870（doi：10.1098 / rsos.181870）），并提出了一个三方面的概念框架来描述与生物物理模型相关的直接，间接和上下文不确定性。其次，我们反思该框架的开发过程，以强调成功合作所面临的挑战以及跨学科深入合作的重要性。这包括解决经常不平等的纪律处分以及早期及早合作问题框架的需求。我们建议，这两种见解对于不确定性建模领域的研究人员和从业人员，以及对跨学科环境研究感兴趣的人员，都具有补充和有益的意义。文章最后总结了跨学科研究的局限性以及向跨学科性的转变，其中也包括非科学家。