#L04 System Engineering for ALM

Abstract

The evolution of a climate model, considered as a whole or as a collection of components, takes place through an iterative process of conceptualization, design, development, application, analysis, and re-conceptualization. It is possible for this evolution to proceed "organically", depending on the changing passions and proclivities of the group of developers that happen to be engaged over time. It is also possible to conceive a more mission-oriented approach that takes explicit account of the objectives of various stakeholders, including science drivers but potentially also policy applications. Here I use the ACME Land Model as an example to illustrate how a System Engineering approach might be applied to the iterative model development process to help identify requirements, guide designs, quantify risks and uncertainties, and deliver mission-specific outcomes for stakeholders. I examine the extent to which the current ACME process already follows a System Engineering approach, and outline some suggestions for how modest recasting of processes could result in a higher likelihood for successful mission outcomes (meeting stakeholder needs). A central recommendation is to consider the model as a system of systems, where a common process for requirements gathering, design, prototyping, implementation, deployment, and analysis can be applied recursively at multiple levels in the system hierarchy. This approach borrows extensively from lessons learned by NASA engineering centers through their experience with multi-component missions with deep system hierarchies. Examples for ALM provided here could serve as a springboard for discussion in relation to other ACME components, and to the ACME coupled system as a whole.