Open Collections

Tsembel, Jangar

University of British Columbia

The cause of extreme events, such as floods, is often attributed to global warming, overlooking the effects of local drivers induced by human activity. To examine the effect of clearcut logging on peak flows, this study draws heavily on attribution science and the science of causation using recent developments in nonstationary frequency analyses. Small opportunistic control (8.2 km²) and treatment (6.3 km²) watersheds in the snow environment of British Columbia’s interior, located 74 km apart, were paired in the magnitude-frequency domain furthering its applicability to forest hydrology. The peak flow frequency distribution was allowed to change over time by enabling its parameters to change in accordance with climate and land cover covariates. The study revealed that clearcut logging in 43% of the watershed area increased the mean (+46%) and standard deviation (+77%) of peak flows. Consequently, the 20-yr, 50-yr, and 100-yr peak flow events became 7, 10, and 14 times more frequent. Nonstationary modelling indicated larger events having a larger effect on peak flow magnitude and frequency, caused by clearcut logging in a watershed with an inherently mild and concave peak flow frequency curve, typical of snowmelt-driven flood regimes. The temporal and spatial distribution of clearcuts, within the critical mid-elevation zone including south-facing slopes, further accentuated the sensitivity of peak flow response to high cut levels. While snowmelt from cutblocks and roads increased the mean of peak flows, the amplified increase in variability was hypothesized to be caused by synchronized melt from logged and unlogged areas. The study compared conventional methods to a novel approach illustrating the efficacy of causal over non-causal experiments, urging more use of a probabilistic framework in studies of forests and flood relations in watersheds across all sizes and hydroclimate regimes.

Text

2025-01-31

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/83667

Forestry

University of British Columbia

UBCV

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