2016-01-26_REVISION - C1981014 - Laserfiche WebLink

Home Browse Search

Proper use of model stratification would be to divide the entire permit area into sections based on similar slope and cover, reclaimed, non -mined, then generate model output, then compare and perhaps average model output. One assumption of the RUSLE model is the output generated (a simple multiplication product from 5 variables) would be greater, equal, or less than the actual sediment measured from the RDA. Model uncertainty increases greatly for soil loss values less than 1 ton/acre/year and can be 500% in error in certain conditions. Nothing in the sediment report identified model uncertainty, accuracy and precision, or provided any validation of the output generated to actual erosion or observed sediment deposition. The January 13, 2016 sediment report failed to demonstrate annual erosion rate accurately nor does it provide any comparison of existing reclaimed condition and non -mined condition. The RUSLE model output only demonstrated a slight change in vegetation canopy for the exact same land conditions. These results demonstrate only a slight change in vegetation canopy cover and more accurately represent a very basic temporal sediment analysis, not a spatial analysis of reclaimed and non -mined lands comparison as implied. The Addendum (1/8/2016) only presents model output for less than half of the mine site. The Addendum also speculates, with no supporting information, that averaging RUSLE model output from three sediment demonstrations is likely the most accurate over the long-term. It must be noted that the RUSLE model output was not validated by any method and has output values generated (not measured) ranging from 0.275 to 0.564. The output values are too small and similar to declare differences or that one area is better when compared to the other, especially when model variability and error are considered. I believe the demonstration with only RUSLE output with a single variable or with two variables over a subset of the permit area greatly biases and simplifies the output; instead, model efforts should be applied with better stratification of the watershed within the entire permit area. There are better, newer technology and more powerful hydrology models like WEPP, SWAT, and others that can better divide the permit area to be modeled into distinct hillslopes or sub -watersheds and potentially identify areas that are contributing greatly to the sediment movement and runoff. I suspect the current sediment analysis demonstrations from 2016 report have used the 1997 version of RUSLE. Erosion along County Road 92 outside the permit area (on either side of the green gate that spans County Road 92) was very obvious to all who attended the November 11, 2015 field visit. Surface runoff, erosion, infiltration, and sediment deposition are hydrological processes that occur infrequently and generally over very short time periods and are rarely observed while occurring. This makes determination of location of accelerated erosion and sediment sources, particularly nonpoint source erosion, difficult to observe and difficult to determine the direct and cumulative impacts that occur over time. Statements that sediment erosion and deposition near the green gate have nothing to do with EFCI's getting its bond released and are not EFCI's responsibility because they are outside the permit boundary are clearly incorrect. The cause of this obvious erosion and sediment deposition that was observed just outside the permit boundary was primarily generated within the EFCI permit boundary on disturbed and reclaimed land, undisturbed land, and a recently disturbed road. Parts of County Road 92 within the permit boundary are frequently graded, have water bars and soil berms that are frequently created or repaired, and some ditch maintenance that is frequently performed. The road area