NATURAL DAMS AND FLOODS OF LEGEND AT MOUNT BAKER VOLCANO--EVIDENCE FROM VOLCANIC STRATIGRAPHY OF THE SHERMAN CRATER ERUPTIVE PERIOD (AD 1843 TO PRESENT)

Initial products of the Sherman Crater eruptive period were the tephra set YP in 1843 and the multi-branched Morovitz Creek Lahar in 1845-1847. Deposits were distributed over much of the SE flank of the volcano, drained by the Baker River, a major tributary of the regional Skagit River. The pale-colored, altered-lithic ash contributed to the unique description, by pioneer paleohydrologist JE Stewart after 1910, of the "tree-staining flood of 1856" in the Skagit River. The dates and approximate magnitudes of that flood and an even larger "Indian-legend flood of 1815" were based by Stewart on the memories, recorded by settlers in 1879 and later, of elderly Native Americans from their youth.

Another legend-based account attributed a large pre-settlement (pre-1859) flood on the Skagit River to failure of a landslide dam of the Baker River. This account probably relates to damming of the river at one of two places by the Morovitz Creek lahar. Our detailed Holocene stratigraphy, and interpretations by pioneer miner/geologist J Morovits, show the maximum height of any blockage was < 4, m, not the > 25 m of legend, and that erosion of at least one of the two blockages was gradual. No significant flood on the Skagit River resulted by breaching of a natural dam on the Baker River during the 19th Century. Natural-dam or log-jam failure elsewhere on the Skagit may have contributed to the account of a rapid rise of the largest flood of legend.

We redate Stewart's flood of 1856 as between 1843 (the tephra age) and 1850. That minimum date is based on re-evaluation of dendrochronology--Stewart's interpretation of a 62-yr-old fir tree cut from a flood bar in 1922. He assumed an ecesis period of 4 yrs (yielding 1856), whereas a minimum ecesis period on fluvial gravel in the area of >10 yrs (yielding pre-1850) is a conservative estimate.

All the 19th Century legends of pre-settlement flooding can be explained by a single, probably rainfall-runoff flood between 1843 and 1850. No account is of more than one flood, and the earliest and best known account (1879) describes "the one flood of Indian legend." We also doubt a large 1815 flood because of excessive antiquity given other Native American accounts, in particular the example of what we believe to be the 1843 eruption of Sherman Crater. That eruption was interpreted in 1877 and later as occurring ca. 1810-1820.