We've Moved!
Visit SDSU’s new digital collections website at https://digitalcollections.sdsu.edu
Description
Wildfire is a common and natural cause of landcover change in Southern California. Wildfire has the potential to impact streamflow regimes through changes in plant evaporation and soil hydrophobicity. Streamflow regimes can be depicted through the use of the Flow Duration Curve (FDC). This study aimed to investigate the impact of fire on FDCs in four Southern California watersheds in the context of a paired watershed approach. Due to soil hydrophobicity and changes in plant evaporation, it was hypothesized that all FDC percentile flows would increase within the first five years after fire. Due to the rapid recovery of shrubland vegetation, it was hypothesized that the impact of fire on the FDC will decrease progressively with post-fire vegetation age within the first 10 years after fire. Two contrasting methods were utilized to assess the impact of fire on FDCs. The Regression Modeling Approach (RMA) consisted of developing a linear regression relationship between flow percentiles of watershed pairs during a pre-fire calibration period and then used to determine how flow percentiles change after the fire occurred. The Parametric Modeling Approach (PMA) utilized a mathematical expression to describe the observed FDC and tested the expressions' parameters for trends after the fire had occurred. The quality of model calibration for both methods was largely dependent on the shape of the observed FDC and the non-zero flow percentage (NZFP). Five years after fire, the RMA results showed FDC percentile flows from 8% to 100% increased in magnitude with the most significant increases from 8% to 35%. Ten years after fire, changes to FDC percentile flows persisted. The PMA resulted in statistically significant trends in one model parameter (representing NZFP) for two of the four test watersheds. Variations in model parameter values tended to follow variations in annual rainfall totals. After fire, NZFPs were consistently larger relative to prefire NZFPs with similar annual rainfall totals. The results from both methods demonstrated that the impacts of fire on the FDC persisted longer than hypothesized and that changes caused by reductions in plant evaporation were more important than those caused by soil hydrophobicity.
