Western Oregon Wet Dry (WOWTDR) annual predictions of late summer streamflow status for western Oregon, 2019-2021

The Western Oregon WeT DRy (WOWTDR) model provides decision makers with a spatially explicit map of western Oregon streams that are predicted to be either wet or dry in late summer. This data publication includes all of the input and output files for producing these predictions with the WOWTDR model. The model output presented here is the mean of WOWTDR predictions for the 2019-2021 calendar years for the purpose of providing a single prediction of wet or dry for each stream reach. Predictions years correspond with observation years of the model calibration data. This model was trained using data collected with the Flow Permanence (FLOwPER) application. This data release provides the WOWTDR model and WOWTDR predictions on streamlines across 426 HUC12s in western Oregon, with the intent to characterize a process for developing WOWTDR predictions. Also included is all R code and Python code needed to run and process this model.
The purpose of the Western Oregon WeT DRy (WOWTDR) model is to predict a “wet” or “dry” classification for each stream reach in western Oregon for which supporting LiDAR data are available, as well as provide an estimate of certainty for that classification, in support of planning for perennial streams at multiple scales. Results are intended to support forest retention buffer estimates and targeted field collection surveys. The model is also designed to support project-specific planning with a sense of what is known and unknown by providing existing flow permanence observations where predictions are more certain and where predictions are ambiguous, with the latter being useful for targeting field surveys. Additionally, the WOWTDR provides a representation of where the model is uninformed (indicated by “Extrapolation = TRUE”), which can be used to focus field observation campaigns to validate and improve future iterations of the model by feeding the FLOw PERmanence (FLOwPER) application database.
For more information about this model and these data, see Burnett et al. (in review).