Open Collections

Understanding and predicting forest harvesting effects on peak flows in snow environments with nonstationary frequency modelling

University of British Columbia

A century of paired watershed studies evaluated the effects of forest harvesting on peak flows by pairing events by equal chronology. This method has recently come under repeated criticism and calls have been made to abandon the practice and pair by equal frequency instead. However, the stationarity assumption imposed by conventional frequency analyses complicates the use of frequency pairing because peak flows contain change-point and trend-shifting nonstationarities caused by continuous harvesting and forest regrowth. Here, a new nonstationary frequency pairing method was introduced to evaluate harvesting effects by allowing the parameters of peak flow frequency distributions to change in time using physically-based covariates. This research falls within the emerging field of “attribution science”, which uses observations and models to identify separately the factors contributing to extremes. The outcomes of applying this new method at five treatment-control, snow-dominated watersheds (37 km2 to 3550 km2) revealed how harvesting increased the magnitude and frequency of not only small (< 10-yr return period) but also large (>10-yr return period) peak flows. This was a consequence of changes to both the means (+28% to +113%) and standard deviations (no effects to +110%) of the peak flow frequency distributions. Large peak flows became 3-4 (10-20) times more frequent in the least (most) sensitive watershed. The different treatment effects reveal contrasting watershed sensitivities to harvesting, owing to different physiographic characteristics and logging histories. Based on the collective outcomes, a physical model encapsulating harvesting effects at the stand- and watershed-levels was developed to advance the probabilistic understanding of the forests and floods relation. Advantages of the new method include: (i) relaxing the watershed size and proximity constraints between control and treatment watersheds; (ii) detecting small levels of change in the peak flow time series with moderate harvesting levels; (iii) bypassing the need for a calibration equation, hence eliminating associated sources of uncertainty; (iv) making use of larger sample sizes to conduct frequency analysis, which make better inferences about the effects on extreme events; and (v) allowing for the estimation of harvesting effects at different historic snapshots of a watershed, thus providing an evaluation of hydrologic recovery.

Text

2022-03-31

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/73836

Forestry

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/