Problem 8: Layers

Dear Editor,

The model used to simulate underground water in the Republican River Basin assumes that all aquifer water is interconnected, even if it is not. It assumes that the aquifer level does not change and that the amount of water released by the aquifer does not change over the decades. The modelers acknowledge that if the water level were permitted to change in the simulation, the model would report that the aquifer had gone dry in some areas, even though it has not in reality.

To overcome this problem, the modelers picked an aquifer level and rate at which it releases water and locked it in those values.

This is a quote from the modelers: "As a single layer transient model, the only aquifer properties required are the hydraulic conductivity and specific yield. However this model encounters severe numerical difficulties when run ... largely due to accidental cell dry-out.

In order to overcome these difficulties, the simulations were run with a fixed saturated thickness and hence fixed transmissivity. Hydraulic conductivities were specified as a discrete set of points (32 points in Nebraska) which were then kriged (spread out) to obtain a value in every cell."

Even though the aquifer level is fixed in the simulation, the modelers also report changes in aquifer levels and report that declines in the aquifer level reduce stream flow and that these declines are caused by irrigation. At the same time, and as we have shown earlier, the Model has a very high error rate in aquifer level predictions.

There is a disconnect between the model documentation and the reports generated by the model.

By comparison, the simulation used on the Platte River uses multiple layers.

Because the model is a single layer simulation, there is little differentiation between water that isn't found until 200 feet below the surface of the earth and water that is within 10 feet. The difference in effect of aquifers that are not close to the stream, either in horizontal or vertical proximity, and those that are is significant.

Steve Smith,

WaterClaim.org

Imperial