About the Watershed
Geomorphology
Erosion in both the uplands and in the creeks of our watershed has many repercussions, including flooding, loss of natural habitat, and threats to homes. What follows is a summary of the Geomorphic Assessment of the Corte Madera Creek Watershed, prepared by Matt Smeltzer, James Reilly, and David Dawdy, a study commissioned by Friends and published in 2000 (GeomorphES-copy.pdf).
Sand and gravel deposition downstream from Ross suggests that Corte Madera Creek has an unusually high sediment yield. Widening channels and bank failures also contribute to the perception that erosion adds significantly to this yield. These deposits reduce flood control effectiveness in the channel. This study estimates bedload sediment at Ross and investigates whether the sediment load is unnaturally high.
Sediment Budget Results
This study estimates that about 7,000 tons/year of bedload are delivered to Ross, or about 450 tons/square mile/year. Due to persistent upland land use impacts, namely increased drainage density caused by 19th century logging and grazing, the Corte Madera Creek watershed’s bedload sediment yield is unnaturally high.
Sediment Sources in the Watershed
This study’s sediment budget estimated that channel bed and bank erosion in the watershed’s alluvial channel network generated about 9 percent of the total bedload sediment load at Ross, for 1976–1999. Fluvial transport from upland channel networks generated about 91 percent of the total sediment yield at Ross. This 91:9 ratio of upland/channel bank sediment sources is comparable to results of detailed sediment budgets compiled for other Marin County watersheds.
Total elimination of bank erosion and systemic channel widening throughout the alluvial channel network would probably reduce bedload sediment delivery to Ross by as much as about 430 tons/yr, only 6 percent of the total bedload delivered to Ross. Total elimination of the additional sediment supply by dealing with problem sediment sources and improved hillslope management practices would probably reduce bedload sediment delivered to Ross by as much as about 1,600 tons/yr, or about 20% of the annual bedload inflow.
This study indicates that the San Anselmo Creek and Sleepy Hollow Creek subwatersheds contribute about 29 percent and 26 percent, respectively, of the total bedload sediment inflow at Ross.
Present Trajectory of Channel Change
This study also evaluated the present state and trajectory of the channel’s natural geomorphic recovery from recent channel entrenchment. Corte Madera Creek’s alluvial channel network became moderately to deeply entrenched in the Holocene valley fill in about 1850–1910, abandoning its floodplain. Rapid channel entrenchment was evidently in partial response to logging and increasing livestock grazing intensity from the middle to late 1800s, coinciding with a period of somewhat greater than normal precipitation. After about 1910, numerous natural bedrock and human structures, like culverts and small dams to create ponds, outcropped in the channel bed; these slowed the channel incision rate and accelerated channel widening. Natural geomorphic recovery processes that restore aquatic and riparian habitat lost during channel entrenchment are operating in the Corte Madera Creek watershed, including progressive upstream channel aggradation in the lower portion of the mainstem Corte Madera Creek and, in the middle and upper portions of the alluvial channel network, channel bed level stabilization, channel widening, inset floodplain formation, and pool-riffle development.
Although progressive upstream channel aggradation evidently ceased in about 1964, constraints imposed by urbanization of the pre-entrenchment floodplain limit the rate of natural habitat improvement both by preventing channel widening with bank protection and flood control structures, and routing storm water directly into the channel network from impermeable surfaces.
As a priority, projects intended to improve flood control and/or aquatic and riparian habitat, habitat-supporting processes, and flood control should seek opportunities to increase active channel width rather than strictly prevent bed incision or bank retreat. Implementing site stormwater retention/drainage best management practices (referred to as Low Impact Development, or LID) that would increase alluvial groundwater storage and summer low flow discharges in the watershed would be beneficial salmonid habitat. LID also improves water quality in the creeks.
