DWR_2726118 - Laserfiche WebLink

 Page 24 Upper and Lower Emerald Valley Dam Failure Report January 23, 2015 8.2 Volumetric Analysis Based on results of no breach scenario, a total volume of storm was 456 acre-feet, received primarily during a period of about 48 hours. Three distinct peaks were observed: 1. 12 SEPT, 05:10, ~228 cfs 2. 12 SEPT, 13:45, ~475 cfs 3. 12 SEPT, 23:00, ~513 cfs Breach analysis indicates the breach of UEV created peak discharge of 669 cfs at Keeton Spillway, which was increase of about 194 cfs from what would have been observed with no dams in place. However, the duration of increase in the system from the failure of UEV and LEV was only about 10 minutes for each dams’ failure. A subsequent peak discharge on Thursday night reached nearly the same discharge, but the duration was much longer due to increased storm runoff volume. The volume of water available from both UEV and LEV dams was only 3.7% of total volume of runoff generated during the entire September 2013 storm event. 9.0 Conclusions 9.1 Possible Causes of Failure The data presented herein including rainfall data, eyewitness accounts, and hydrologic modeling clearly show the Emerald Valley Dams were overtopped and failed primarily due to backward headcutting erosion during the overtopping processes. From a hydrologic adequacy perspective of the failed dams, little discussion will be provided other than to note the dams had pipe emergency spillways, which are not allowed in accordance with DSB Rules and Regulations for Dam Safety and Dam Construction. Additionally, it should be noted that the dam crests were not level, with numerous irregularities and variations in elevation. The crests were narrow with rounded shoulders and had no defined cross slope to the upstream or the downstream directions. This allowed overtopping flows to concentrate in the low spots of the dam crest and expedite the headcutting processes. From another perspective, the forensic review of the failed dams showed they presented numerous concerns with multiple pipe penetrations and no filter zones or other modern dam design approaches to reduce risk of backward erosion piping or failure due to conduit deterioration. Based on anecdotal reports, Lower Emerald Valley Dam failed in 1997 and was “rebuilt” at that time. The 1997 breach was also located at the left abutment. A review of aerial photographs between 1997 and 2013 shows that the dam in this zone was not reconstructed to match the relatively shallow slopes of the rest of the right portion of the dam. The photos indicate the cross section in this area was much narrower and actually had several large CMP pipe penetrations. Given this, it is also likely little earthwork control/compaction was given during the 1997 “reconstruction” of LEV. The narrow cross section, multiple pipe penetrations, and likely poorly compacted soils would A) increase probability of failure during overtopping and B) present possibility of a combined failure modes of overtopping and backward erosion piping due to increased hydraulic gradient both during overtopping event and narrowed cross section of dam.