A dynamic statistical experiment for atmospheric interactions

Interactions among atmospheric parameters exist at different scales. The pristine approach for observational or model data analysis involves changing the input parameters one at a time (OAT) and studying the effect on the system. Limitations of this approach for atmospheric applications are discussed. A fractional factorial (FF) based study is evolved and a methodology is outlined involving dynamic graphical analysis. Observational data from the FIFE and HAPEX‐MOBILHY experiments are utilized with a vegetation and soil moisture scheme dynamically coupled in a planetary boundary layer model to demonstrate the robustness of this approach. Both low‐resolution and high‐resolution designs are considered. Various aspects of the vegetation‐atmosphere interactions are delineated. Results obtained from the interaction‐based FF approach differ considerably from the earlier OAT‐type studies.