Skip to main content
ARS Home » Pacific West Area » Boise, Idaho » Northwest Watershed Research Center » Research » Publications at this Location » Publication #84069
Title: SPATIAL VARIABILITY OF FROZEN SOIL RUNOFF AT DIFFERENT SCALES

Author
item Seyfried, Mark
item Flerchinger, Gerald
item MURDOCK, MARK - UNIVERSITY OF IDAHO

Submitted to: International Symposium on Physics Chemistry and Ecology of Seasonally Froz
Publication Type: Proceedings
Publication Acceptance Date: 10/10/1997
Publication Date: N/A
Citation: N/A

Interpretive Summary: Soil freezing can cause dramatic reductions in infiltration rates, which have been associated with major flooding and erosion events in northern latitudes. In order to predict these events we need to be extend the scientific research that has been conducted at relatively small scales to very much larger watersheds. The chief difficulty with this is the high degree of spatial variability found under natural field conditions. We set up an experiment to measure frozen soil runoff at different scales in order to determine what measurement size is optimal. We found that snow depth variability, not soil temperature or water content, controlled the spatial varaibility of runoff generation. Snow depth was highly variable at 1 m measurement scales but "averaged out" at 30 m2 plots.

Technical Abstract: Soil freezing can cause dramatic reductions in infiltration rates, which have been associated with major flooding and erosion events in northern latitudes. In this paper we analyzed field runoff data collected at scales ranging from 1-m2 to 13 ha with the intent of identifying practical scales for modeling frozen soil runoff. Observed runoff variability was then linked to landscape variables to determine what the source of variability is at different scales. Measured variables included soil water content, soil temperature and snow depth. We found that: (i) there is extremely high spatial variability of runoff at the 1 m2 scale which is probably due to the highly variable snow distribution affected by vegetation, (ii) this variability is effectively averaged out at the scale of 30-m2 plots, and (iii) an increase in scale introduces a new source of variability, topography, which must be accounted for. It may be that at larger scales topographic effects are also averaged out.