Describing and Testing Nonlinear Treatment Effects in Paired Watershed Experiments
In forest hydrology research, treatment effects are commonly assumed to be proportional to the hydrologic response measured on a control watershed, and are fitted and tested for significance by linear regression techniques. More likely the treatment effect (measured as the difference between pre- and posttreatment regressions of the treatment basin's response on the control basin's response to weather) is nonlinear, unimodal in midrange, and asymptotically zero in one or both extreme ranges. This paper offers simple models for describing such treatment effects and gives illustrative fitting and testing procedures. Forest Sci. 30:305-313.
No Supplementary Data
No Article Media