Ringold, P.L., M. Bollman, J. Van Sickle, J. Barker, and J. Welty. 2000. Predictions of stream wood recruitment from riparian forests: Effects of data resolution. Pages 505-510 in P.J. Wigington, Jr., and R.L. Beschta, editors, Riparian Ecology and Management in Multi-Land Use Watersheds, Proc. AWRA Specialty Conference, Portland, OR, Aug. 28-31, 2000.
We evaluate whether different levels of detail of riparian forest characterizations result in different predictions of stream wood recruitment from riparian forests in northwestern Oregon. If less detailed information provides the same estimate of this function as more detailed information, then data can be acquired with a lower level of detail and cost. This analysis is based on field observations at 110 forested riparian plots in northwestern Oregon with measurements at each including the size, location, and species of each tree. This field information is represented with two levels of spatial resolution and with or without stand classification. These representations serve as input to a model that predicts the amount of stream wood from each plot. Changes in the input data representations can change the amount of stream wood predicted to be derived from an individual plot by an order of magnitude, although the median change tends to be small. We also consider the change in prediction for the total loading to the stream network. In this network case, the change in stream wood prediction is generally small. Thus, less detailed data may prove acceptable for assessment of stream wood loading to a stream network.